# EC2 Fleet and Spot Fleet EC2 Fleet and Spot Fleet are designed to be a useful way to launch a fleet of tens, hundreds, or thousands of Amazon EC2 instances in a single operation. Each instance in a fleet is either configured by a [launch template](ec2-launch-templates.md) or a set of launch parameters that you configure manually at launch. **Topics** + [ ## Features and benefits ](#ec2-fleet-features-and-benefits) + [ # Which is the best fleet method to use? ](which-fleet-method-to-use.md) + [ # Configuration options for your EC2 Fleet or Spot Fleet ](ec2-fleet-configuration-strategies.md) + [ # Work with EC2 Fleet ](manage-ec2-fleet.md) + [ # Work with Spot Fleet ](work-with-spot-fleets.md) + [ # Monitor your EC2 Fleet or Spot Fleet ](fleet-monitor.md) + [ # Tutorials for EC2 Fleet ](fleet-tutorials.md) + [ # Example CLI configurations for EC2 Fleet ](ec2-fleet-examples.md) + [ # Example CLI configurations Spot Fleet ](spot-fleet-examples.md) + [ # Quotas for EC2 Fleet and Spot Fleet ](fleet-quotas.md) ## Features and benefits Fleets provide the following features and benefits, enabling you to maximize cost savings and optimize availability and performance when running applications on multiple EC2 instances. **Multiple instance types** A fleet can launch multiple instance types, ensuring it isn't dependent on the availability of any single instance type. This increases the overall availability of instances in your fleet. **Distributing instances across Availability Zones** A fleet can launch into multiple Availability Zones, enabling you to reduce costs and improve availability. If your fleet includes Spot Instances, the fleet automatically selects Availability Zones based on your preferences regarding price and interruptions. **Multiple purchasing options** A fleet can launch multiple purchase options (Spot and On-Demand Instances), allowing you to optimize costs through Spot Instance usage. You can also take advantage of Reserved Instance and Savings Plans discounts by using them in conjunction with On-Demand Instances in the fleet. **Automated replacement of Spot Instances** If your fleet includes Spot Instances, it can automatically request replacement Spot capacity if your Spot Instances are interrupted. Through [Capacity Rebalancing](ec2-fleet-capacity-rebalance.md), a fleet can also monitor and proactively replace your Spot Instances that are at an elevated risk of interruption. **Reserve On-Demand capacity** A fleet can use an [On-Demand Capacity Reservation](ec2-fleet-on-demand-capacity-reservations.md) to reserve On-Demand capacity. A fleet can also include [Capacity Blocks for ML](ec2-capacity-blocks.md), allowing you to reserve GPU instances on a future date to support short duration machine learning (ML) workloads. # Which is the best fleet method to use? As a general best practice, we recommend launching fleets of Spot and On-Demand Instances with Amazon EC2 Auto Scaling because it provides additional features you can use to manage your fleet. The list of additional features includes automatic health check replacements for both Spot and On-Demand Instances, application-based health checks, and an integration with Elastic Load Balancing to ensure an even distribution of application traffic to your healthy instances. You can also use Auto Scaling groups when you use AWS services such as Amazon ECS, Amazon EKS (self-managed node groups), and Amazon VPC Lattice. For more information, see the [Amazon EC2 Auto Scaling User Guide](https://docs.aws.amazon.com/autoscaling/ec2/userguide/). If you can't use Amazon EC2 Auto Scaling, then you might consider using EC2 Fleet or Spot Fleet. EC2 Fleet and Spot Fleet offer the same core functionality. However, EC2 Fleet is only available using a command line and does not provide console support. Spot Fleet provides console support, but is based on a legacy API with no planned investment. Use the following table to determine which fleet method to use. **** | Fleet method | When to use? | Use case | | --- | --- | --- | | [Amazon EC2 Auto Scaling](https://docs.aws.amazon.com/autoscaling/ec2/userguide/what-is-amazon-ec2-auto-scaling.html) | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/which-fleet-method-to-use.html) | Create an Auto Scaling group that manages the lifecycle of your instances while maintaining the desired number of instances. Supports horizontal scaling (adding more instances) between specified minimum and maximum limits. | | [EC2 Fleet](manage-ec2-fleet.md) | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/which-fleet-method-to-use.html) | Create an `instant` fleet of both On-Demand Instances and Spot Instances in a single operation, with multiple launch specifications that vary by instance type, AMI, Availability Zone, or subnet. The Spot Instance allocation strategy defaults to `lowest-price` per unit, but we recommend changing it to `price-capacity-optimized`. | | [Spot Fleet](work-with-spot-fleets.md) | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/which-fleet-method-to-use.html) | Use Spot Fleet only if you need console support for a use case for when you would use EC2 Fleet. | # Configuration options for your EC2 Fleet or Spot Fleet When planning your EC2 Fleet or Spot Fleet, we recommend that you consider the following options when deciding how to configure your fleet. **** | Configuration option | Question | Documentation | | --- | --- | --- | | Fleet request type | Do you want a fleet that submits a one-time request for the desired target capacity, or a fleet that maintains target capacity over time? | [EC2 Fleet and Spot Fleet request types](ec2-fleet-request-type.md) | | Spot Instances | Do you plan to include Spot Instances in your fleet? Review the Spot best practices and use them when you plan your fleet so that you can provision the instances at the lowest possible price. | [Best practices for Amazon EC2 Spot](spot-best-practices.md) | | Spending limit for your fleet | Do you want to limit how much you'll pay for your fleet per hour? | [Set a spending limit for your EC2 Fleet or Spot Fleet](ec2-fleet-control-spending.md) | | Instance types and attribute-based instance type selection | Do you want to specify the instance types in your fleet, or let Amazon EC2 select the instance types that meet your application requirements? | [Specify attributes for instance type selection for EC2 Fleet or Spot Fleet](ec2-fleet-attribute-based-instance-type-selection.md) | | Instance weighting | Do you want to assign weights to each instance type to represent their compute capacity and performance, so that Amazon EC2 can select any combination of available instance types to fulfil your desired target capacity? | [Use instance weighting to manage cost and performance of your EC2 Fleet or Spot Fleet](ec2-fleet-instance-weighting.md) | | Allocation strategies | Do you want to decide whether to optimize for available capacity, price, or instance types to use for the Spot Instances and On-Demand Instances in your fleet? | [Use allocation strategies to determine how EC2 Fleet or Spot Fleet fulfills Spot and On-Demand capacity](ec2-fleet-allocation-strategy.md) | | Capacity Rebalancing | Do you want your fleet to automatically replace at-risk Spot Instances? | [Use Capacity Rebalancing in EC2 Fleet and Spot Fleet to replace at-risk Spot Instances](ec2-fleet-capacity-rebalance.md) | | On-Demand Capacity Reservation | Do you want to reserve capacity for the On-Demand Instances in your fleet? | [Use Capacity Reservations to reserve On-Demand capacity in EC2 Fleet](ec2-fleet-on-demand-capacity-reservations.md) | # EC2 Fleet and Spot Fleet request types The request type for an EC2 Fleet or Spot Fleet determines whether the request is synchronous or asynchronous, and whether it is a one-time request for the desired target capacity or an ongoing effort to maintain the capacity over time. When configuring your fleet, you must specify the request type. Both EC2 Fleet and Spot Fleet offer two request types: `request` and `maintain`. In addition, EC2 Fleet offers a third request type called `instant`.Fleet request types `instant` (EC2 Fleet only) If you configure the request type as `instant`, EC2 Fleet places a synchronous one-time request for your desired capacity. In the API response, it returns the instances that launched and provides errors for those instances that could not be launched. For more information, see [Configure an EC2 Fleet of type instant](instant-fleet.md). `request` If you configure the request type as `request`, the fleet places an asynchronous one-time request for your desired capacity. If capacity diminishes due to Spot interruptions, the fleet does not attempt to replenish Spot Instances, nor does it submit requests in alternative Spot capacity pools if capacity is unavailable. When creating a Spot Fleet of type `request` using the console, clear the **Maintain target capacity** checkbox. `maintain` (default) If you configure the request type as `maintain`, the fleet places an asynchronous request for your desired capacity, and maintains it by automatically replenishing any interrupted Spot Instances. When creating a Spot Fleet of type `maintain` using the console, select the **Maintain target capacity** checkbox # Configure an EC2 Fleet of type instant The EC2 Fleet of type *instant* is a synchronous one-time request that makes only one attempt to launch your desired capacity. The API response lists the instances that launched, along with errors for those instances that could not be launched. There are several benefits to using an EC2 Fleet of type *instant*, which are described in this article. Example configurations are provided at the end of the article. For workloads that need a launch-only API to launch EC2 instances, you can use the RunInstances API. However, with RunInstances, you can only launch On-Demand Instances or Spot Instances, but not both in the same request. Furthermore, when you use RunInstances to launch Spot Instances, your Spot Instance request is limited to one instance type and one Availability Zone. This targets a single Spot capacity pool (a set of unused instances with the same instance type and Availability Zone). If the Spot capacity pool does not have sufficient Spot Instance capacity for your request, the RunInstances call fails. Instead of using RunInstances to launch Spot Instances, we recommend that you rather use the CreateFleet API with the `type` parameter set to `instant` for the following benefits: + **Launch On-Demand Instances and Spot Instances in one request.** An EC2 Fleet can launch On-Demand Instances, Spot Instances, or both. The request for Spot Instances is fulfilled if there is available capacity and the maximum price per hour for your request exceeds the Spot price. + **Increase the availability of Spot Instances.** By using an EC2 Fleet of type `instant`, you can launch Spot Instances following [Spot best practices](spot-best-practices.md) with the resulting benefits: + **Spot best practice: Be flexible about instance types and Availability Zones.** Benefit: By specifying several instance types and Availability Zones, you increase the number of Spot capacity pools. This gives the Spot service a better chance of finding and allocating your desired Spot compute capacity. A good rule of thumb is to be flexible across at least 10 instance types for each workload and make sure that all Availability Zones are configured for use in your VPC. + **Spot best practice: Use the price-capacity-optimized allocation strategy.** Benefit: The `price-capacity-optimized` allocation strategy identifies instances from the most-available Spot capacity pools, and then automatically provisions instances from the lowest priced of these pools. Because your Spot Instance capacity is sourced from pools with optimal capacity, this decreases the possibility that your Spot Instances will be interrupted when Amazon EC2 needs the capacity back. + **Get access to a wider set of capabilities.** For workloads that need a launch-only API, and where you prefer to manage the lifecycle of your instance rather than let EC2 Fleet manage it for you, use the EC2 Fleet of type `instant` instead of the [RunInstances](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_RunInstances.html) API. EC2 Fleet provides a wider set of capabilities than RunInstances, as demonstrated in the following examples. For all other workloads, you should use Amazon EC2 Auto Scaling because it supplies a more comprehensive feature set for a wide variety of workloads, like ELB-backed applications, containerized workloads, and queue processing jobs. You can use EC2 Fleet of type *instant* to launch instances into Capacity Blocks. For more information, see [Tutorial: Configure your EC2 Fleet to launch instances into Capacity Blocks](ec2-fleet-launch-instances-capacity-blocks-walkthrough.md). AWS services like Amazon EC2 Auto Scaling and Amazon EMR use EC2 Fleet of type *instant* to launch EC2 instances. ## Prerequisites for EC2 Fleet of type instant For the prerequisites for creating an EC2 Fleet, see [EC2 Fleet prerequisites](ec2-fleet-prerequisites.md). ## How instant EC2 Fleet works When working with an EC2 Fleet of type `instant`, the sequence of events is as follows: 1. **Configure:** Configure the [CreateFleet](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_CreateFleet.html) request type as `instant`. For more information, see [Create an EC2 Fleet](create-ec2-fleet.md). Note that after you make the API call, you can't modify it. 1. **Request:** When you make the API call, Amazon EC2 places a synchronous one-time request for your desired capacity. 1. **Response:** The API response lists the instances that launched, along with errors for those instances that could not be launched. 1. **Describe:** You can describe your EC2 Fleet, list the instances associated with your EC2 Fleet, and view the history of your EC2 Fleet. 1. **Terminate instances:** You can terminate the instances at any time. 1. **Delete fleet request:** The fleet request can be deleted either manually or automatically: + Manual: You can [delete the fleet request](delete-fleet.md) after your instances launch. Note that a deleted `instant` fleet with running instances is not supported. When you delete an `instant` fleet, Amazon EC2 automatically terminates all its instances. For fleets with more than 1000 instances, the deletion request might fail. If your fleet has more than 1000 instances, first terminate most of the instances manually, leaving 1000 or fewer. Then delete the fleet, and the remaining instances will be terminated automatically. + Automatic: Amazon EC2 deletes the fleet request some time after either: + All the instances are terminated. + The fleet fails to launch any instances. ## Examples The following examples show how to use EC2 Fleet of type `instant` for different use cases. For more information about using the EC2 CreateFleet API parameters, see [CreateFleet](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_CreateFleet.html) in the *Amazon EC2 API Reference*. **Topics** + [ ### Example 1: Launch Spot Instances with the capacity-optimized allocation strategy ](#instant-fleet-example-1) + [ ### Example 2: Launch a single Spot Instance with the capacity-optimized allocation strategy ](#instant-fleet-example-2) + [ ### Example 3: Launch Spot Instances using instance weighting ](#instant-fleet-example-3) + [ ### Example 4: Launch Spot Instances within single Availability Zone ](#instant-fleet-example-4) + [ ### Example 5: Launch Spot Instances of single instance type within single Availability zone ](#instant-fleet-example-5) + [ ### Example 6: Launch Spot Instances only if minimum target capacity can be launched ](#instant-fleet-example-6) + [ ### Example 7: Launch Spot Instances only if minimum target capacity can be launched of same Instance Type in a single Availability Zone ](#instant-fleet-example-7) + [ ### Example 8: Launch instances with multiple Launch Templates ](#instant-fleet-example-8) + [ ### Example 9: Launch Spot Instance with a base of On-Demand Instances ](#instant-fleet-example-9) + [ ### Example 10: Launch Spot Instances using capacity-optimized allocation strategy with a base of On-Demand Instances using Capacity Reservations and the prioritized allocation strategy ](#instant-fleet-example-10) + [ ### Example 11: Launch Spot Instances using capacity-optimized-prioritized allocation strategy ](#instant-fleet-example-11) + [ ### Example 12: Specify a Systems Manager parameter instead of an AMI ID ](#instant-fleet-example-12) ### Example 1: Launch Spot Instances with the capacity-optimized allocation strategy The following example specifies the parameters required in an EC2 Fleet of type `instant`: a launch template, target capacity, default purchasing option, and launch template overrides. + The launch template is identified by its launch template name and version number. + The 12 launch template overrides specify 4 different instance types and 3 different subnets, each in a separate Availability Zone. Each instance type and subnet combination defines a Spot capacity pool, resulting in 12 Spot capacity pools. + The target capacity for the fleet is 20 instances. + The default purchasing option is `spot`, which results in the fleet attempting to launch 20 Spot Instances into the Spot capacity pool with optimal capacity for the number of instances that are launching. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-49e41922" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 2: Launch a single Spot Instance with the capacity-optimized allocation strategy You can optimally launch one Spot Instance at a time by making multiple EC2 Fleet API calls of type `instant`, by setting the TotalTargetCapacity to 1. The following example specifies the parameters required in an EC2 Fleet of type instant: a launch template, target capacity, default purchasing option, and launch template overrides. The launch template is identified by its launch template name and version number. The 12 launch template overrides have 4 different instance types and 3 different subnets, each in a separate Availability Zone. The target capacity for the fleet is 1 instance, and the default purchasing option is spot, which results in the fleet attempting to launch a Spot Instance from one of the 12 Spot capacity pools based on the capacity-optimized allocation strategy, to launch a Spot Instance from the most-available capacity pool. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-49e41922" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 1, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 3: Launch Spot Instances using instance weighting The following examples use instance weighting, which means that the price is per unit hour instead of per instance hour. Each launch configuration lists a different instance type and a different weight based on how many units of the workload can run on the instance assuming a unit of the workload requires a 15 GB of memory and 4 vCPUs. For example an m5.xlarge (4 vCPUs and 16 GB of memory) can run one unit and is weighted 1, m5.2xlarge (8 vCPUs and 32 GB of memory) can run 2 units and is weighted 2, and so on. The total target capacity is set to 40 units. The default purchasing option is spot, and the allocation strategy is capacity-optimized, which results in either 40 m5.xlarge (40 divided by 1), 20 m5.2xlarge (40 divided by 2), 10 m5.4xlarge (40 divided by 4), 5 m5.8xlarge (40 divided by 8), or a mix of the instance types with weights adding up to the desired capacity based on the capacity-optimized allocation strategy. For more information, see [Use instance weighting to manage cost and performance of your EC2 Fleet or Spot Fleet](ec2-fleet-instance-weighting.md). ``` { "SpotOptions":{ "AllocationStrategy":"capacity-optimized" }, "LaunchTemplateConfigs":[ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"m5.xlarge", "SubnetId":"subnet-fae8c380", "WeightedCapacity":1 }, { "InstanceType":"m5.xlarge", "SubnetId":"subnet-e7188bab", "WeightedCapacity":1 }, { "InstanceType":"m5.xlarge", "SubnetId":"subnet-49e41922", "WeightedCapacity":1 }, { "InstanceType":"m5.2xlarge", "SubnetId":"subnet-fae8c380", "WeightedCapacity":2 }, { "InstanceType":"m5.2xlarge", "SubnetId":"subnet-e7188bab", "WeightedCapacity":2 }, { "InstanceType":"m5.2xlarge", "SubnetId":"subnet-49e41922", "WeightedCapacity":2 }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-fae8c380", "WeightedCapacity":4 }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-e7188bab", "WeightedCapacity":4 }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-49e41922", "WeightedCapacity":4 }, { "InstanceType":"m5.8xlarge", "SubnetId":"subnet-fae8c380", "WeightedCapacity":8 }, { "InstanceType":"m5.8xlarge", "SubnetId":"subnet-e7188bab", "WeightedCapacity":8 }, { "InstanceType":"m5.8xlarge", "SubnetId":"subnet-49e41922", "WeightedCapacity":8 } ] } ], "TargetCapacitySpecification":{ "TotalTargetCapacity":40, "DefaultTargetCapacityType":"spot" }, "Type":"instant" } ``` ### Example 4: Launch Spot Instances within single Availability Zone You can configure a fleet to launch all instances in a single Availability Zone by setting the Spot options SingleAvailabilityZone to true. The 12 launch template overrides have different instance types and subnets (each in a separate Availability Zone) but the same weighted capacity. The total target capacity is 20 instances, the default purchasing option is spot, and the Spot allocation strategy is capacity-optimized. The EC2 Fleet launches 20 Spot Instances all in a single AZ, from the Spot capacity pool(s) with optimal capacity using the launch specifications. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized", "SingleAvailabilityZone": true }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-49e41922" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 5: Launch Spot Instances of single instance type within single Availability zone You can configure a fleet to launch all instances of the same instance type and in a single Availability Zone by setting the SpotOptions SingleInstanceType to true and SingleAvailabilityZone to true. The 12 launch template overrides have different instance types and subnets (each in a separate Availability Zone) but the same weighted capacity. The total target capacity is 20 instances, the default purchasing option is spot, the Spot allocation strategy is capacity-optimized. The EC2 Fleet launches 20 Spot Instances of the same instance type all in a single AZ from the Spot Instance pool with optimal capacity using the launch specifications. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized", "SingleInstanceType": true, "SingleAvailabilityZone": true }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-49e41922" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 6: Launch Spot Instances only if minimum target capacity can be launched You can configure a fleet to launch instances only if the minimum target capacity can be launched by setting the Spot options MinTargetCapacity to the minimum target capacity you want to launch together. When specifying MinTargetCapacity, you must specify at least one of these parameters: SingleInstanceType or SingleAvailabilityZone. In this example, SingleInstanceType is specified, so that all 20 instances must use the same instance type. The 12 launch template overrides have different instance types and subnets (each in a separate Availability Zone) but the same weighted capacity. The total target capacity and the minimum target capacity are both set to 20 instances, the default purchasing option is spot, and the Spot allocation strategy is capacity-optimized. The EC2 Fleet launches 20 Spot Instances from the Spot capacity pool with optimal capacity using the launch template overrides, only if it can launch all 20 instances at the same time. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized", "SingleInstanceType": true, "MinTargetCapacity": 20 }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-49e41922" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 7: Launch Spot Instances only if minimum target capacity can be launched of same Instance Type in a single Availability Zone You can configure a fleet to launch instances only if the minimum target capacity can be launched with a single instance type in a single Availability Zone by setting the Spot options MinTargetCapacity to the minimum target capacity you want to launch together along with SingleInstanceType and SingleAvailabilityZone options. The 12 launch specifications which override the launch template, have different instance types and subnets (each in a separate Availability Zone) but the same weighted capacity. The total target capacity and the minimum target capacity are both set to 20 instances, the default purchasing option is spot, the Spot allocation strategy is capacity-optimized, the SingleInstanceType is true and SingleAvailabilityZone is true. The EC2 Fleet launches 20 Spot Instances of the same Instance type all in a single AZ from the Spot capacity pool with optimal capacity using the launch specifications, only if it can launch all 20 instances at the same time. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized", "SingleInstanceType": true, "SingleAvailabilityZone": true, "MinTargetCapacity": 20 }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5d.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5.4xlarge", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5d.4xlarge", "SubnetId":"subnet-49e41922" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 8: Launch instances with multiple Launch Templates You can configure a fleet to launch instances with different launch specifications for different instance types or a group of instance types, by specifying multiple launch templates. In this example we want have different EBS volume sizes for different instance types and we have that configured in the launch templates ec2-fleet-lt-4xl, ec2-fleet-lt-9xl and ec2-fleet-lt-18xl. In this example, we are using 3 different launch templates for the 3 instance types based on their size. The launch specification overrides on all the launch templates use instance weights based on the vCPUs on the instance type. The total target capacity is 144 units, the default purchasing option is spot, and the Spot allocation strategy is capacity-optimized. The EC2 Fleet can either launch 9 c5n.4xlarge (144 divided by 16) using the launch template ec2-fleet-4xl or 4 c5n.9xlarge (144 divided by 36) using the launch template ec2-fleet-9xl, or 2 c5n.18xlarge (144 divided by 72) using the launch template ec2-fleet-18xl, or a mix of the instance types with weights adding up to the desired capacity based on the capacity-optimized allocation strategy. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt-18xl", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5n.18xlarge", "SubnetId":"subnet-fae8c380", "WeightedCapacity":72 }, { "InstanceType":"c5n.18xlarge", "SubnetId":"subnet-e7188bab", "WeightedCapacity":72 }, { "InstanceType":"c5n.18xlarge", "SubnetId":"subnet-49e41922", "WeightedCapacity":72 } ] }, { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt-9xl", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5n.9xlarge", "SubnetId":"subnet-fae8c380", "WeightedCapacity":36 }, { "InstanceType":"c5n.9xlarge", "SubnetId":"subnet-e7188bab", "WeightedCapacity":36 }, { "InstanceType":"c5n.9xlarge", "SubnetId":"subnet-49e41922", "WeightedCapacity":36 } ] }, { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt-4xl", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5n.4xlarge", "SubnetId":"subnet-fae8c380", "WeightedCapacity":16 }, { "InstanceType":"c5n.4xlarge", "SubnetId":"subnet-e7188bab", "WeightedCapacity":16 }, { "InstanceType":"c5n.4xlarge", "SubnetId":"subnet-49e41922", "WeightedCapacity":16 } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 144, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 9: Launch Spot Instance with a base of On-Demand Instances The following example specifies the total target capacity of 20 instances for the fleet, and a target capacity of 5 On-Demand Instances. The default purchasing option is spot. The fleet launches 5 On-Demand Instance as specified, but needs to launch 15 more instances to fulfill the total target capacity. The purchasing option for the difference is calculated as TotalTargetCapacity – OnDemandTargetCapacity = DefaultTargetCapacityType, which results in the fleet launching 15 Spot Instances form one of the 12 Spot capacity pools based on the capacity-optimized allocation strategy. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"c5d.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5.large", "SubnetId":"subnet-49e41922" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-fae8c380" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-e7188bab" }, { "InstanceType":"m5d.large", "SubnetId":"subnet-49e41922" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "OnDemandTargetCapacity": 5, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 10: Launch Spot Instances using capacity-optimized allocation strategy with a base of On-Demand Instances using Capacity Reservations and the prioritized allocation strategy You can configure a fleet to use On-Demand Capacity Reservations first when launching a base of On-Demand Instances with the default target capacity type as spot by setting the usage strategy for Capacity Reservations to use-capacity-reservations-first. And if multiple instance pools have unused Capacity Reservations, the chosen On-Demand allocation strategy is applied. In this example, the On-Demand allocation strategy is prioritized. In this example, there are 6 available unused Capacity Reservations. This is less than the fleet's target On-Demand capacity of 10 On-Demand Instances. The account has the following 6 unused Capacity Reservations in 2 pools. The number of Capacity Reservations in each pool is indicated by AvailableInstanceCount. ``` { "CapacityReservationId": "cr-111", "InstanceType": "m5.large", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount": 3, "InstanceMatchCriteria": "open", "State": "active" } { "CapacityReservationId": "cr-222", "InstanceType": "c5.large", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount": 3, "InstanceMatchCriteria": "open", "State": "active" } ``` The following fleet configuration shows only the pertinent configurations for this example. The On-Demand allocation strategy is prioritized, and the usage strategy for Capacity Reservations is use-capacity-reservations-first. The Spot allocation strategy is capacity-optimized. The total target capacity is 20, the On-Demand target capacity is 10, and the default target capacity type is spot. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized" }, "OnDemandOptions":{ "CapacityReservationOptions": { "UsageStrategy": "use-capacity-reservations-first" }, "AllocationStrategy":"prioritized" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.large", "SubnetId":"subnet-fae8c380", "Priority": 1.0 }, { "InstanceType":"c5.large", "SubnetId":"subnet-e7188bab", "Priority": 2.0 }, { "InstanceType":"c5.large", "SubnetId":"subnet-49e41922", "Priority": 3.0 }, { "InstanceType":"c5d.large", "SubnetId":"subnet-fae8c380", "Priority": 4.0 }, { "InstanceType":"c5d.large", "SubnetId":"subnet-e7188bab", "Priority": 5.0 }, { "InstanceType":"c5d.large", "SubnetId":"subnet-49e41922", "Priority": 6.0 }, { "InstanceType":"m5.large", "SubnetId":"subnet-fae8c380", "Priority": 7.0 }, { "InstanceType":"m5.large", "SubnetId":"subnet-e7188bab", "Priority": 8.0 }, { "InstanceType":"m5.large", "SubnetId":"subnet-49e41922", "Priority": 9.0 }, { "InstanceType":"m5d.large", "SubnetId":"subnet-fae8c380", "Priority": 10.0 }, { "InstanceType":"m5d.large", "SubnetId":"subnet-e7188bab", "Priority": 11.0 }, { "InstanceType":"m5d.large", "SubnetId":"subnet-49e41922", "Priority": 12.0 } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "OnDemandTargetCapacity": 10, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` After you create the instant fleet using the preceding configuration, the following 20 instances are launched to meet the target capacity: + 7 c5.large On-Demand Instances in us-east-1a – c5.large in us-east-1a is prioritized first, and there are 3 available unused c5.large Capacity Reservations. The Capacity Reservations are used first to launch 3 On-Demand Instances plus 4 additional On-Demand Instances are launched according to the On-Demand allocation strategy, which is prioritized in this example. + 3 m5.large On-Demand Instances in us-east-1a – m5.large in us-east-1a is prioritized second, and there are 3 available unused m5.large Capacity Reservations. + 10 Spot Instances from one of the 12 Spot capacity pools that has the optimal capacity according to the capacity-optimized allocation strategy. After the fleet is launched, you can run [describe-capacity-reservations](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-capacity-reservations.html) to see how many unused Capacity Reservations are remaining. In this example, you should see the following response, which shows that all of the c5.large and m5.large Capacity Reservations were used. ``` { "CapacityReservationId": "cr-111", "InstanceType": "m5.large", "AvailableInstanceCount": 0 } { "CapacityReservationId": "cr-222", "InstanceType": "c5.large", "AvailableInstanceCount": 0 } ``` ### Example 11: Launch Spot Instances using capacity-optimized-prioritized allocation strategy The following example specifies the parameters required in an EC2 Fleet of type instant: a launch template, target capacity, default purchasing option, and launch template overrides. The launch template is identified by its launch template name and version number. The 12 launch specifications which override the launch template have 4 different instance types with a priority assigned, and 3 different subnets, each in a separate Availability Zone. The target capacity for the fleet is 20 instances, and the default purchasing option is spot, which results in the fleet attempting to launch 20 Spot Instances from one of the 12 Spot capacity pools based on the capacity-optimized-prioritized allocation strategy, which implements priorities on a best-effort basis, but optimizes for capacity first. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized-prioritized" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification":{ "LaunchTemplateName":"ec2-fleet-lt1", "Version":"$Latest" }, "Overrides":[ { "InstanceType":"c5.large", "SubnetId":"subnet-fae8c380", "Priority": 1.0 }, { "InstanceType":"c5.large", "SubnetId":"subnet-e7188bab", "Priority": 1.0 }, { "InstanceType":"c5.large", "SubnetId":"subnet-49e41922", "Priority": 1.0 }, { "InstanceType":"c5d.large", "SubnetId":"subnet-fae8c380", "Priority": 2.0 }, { "InstanceType":"c5d.large", "SubnetId":"subnet-e7188bab", "Priority": 2.0 }, { "InstanceType":"c5d.large", "SubnetId":"subnet-49e41922", "Priority": 2.0 }, { "InstanceType":"m5.large", "SubnetId":"subnet-fae8c380", "Priority": 3.0 }, { "InstanceType":"m5.large", "SubnetId":"subnet-e7188bab", "Priority": 3.0 }, { "InstanceType":"m5.large", "SubnetId":"subnet-49e41922", "Priority": 3.0 }, { "InstanceType":"m5d.large", "SubnetId":"subnet-fae8c380", "Priority": 4.0 }, { "InstanceType":"m5d.large", "SubnetId":"subnet-e7188bab", "Priority": 4.0 }, { "InstanceType":"m5d.large", "SubnetId":"subnet-49e41922", "Priority": 4.0 } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ### Example 12: Specify a Systems Manager parameter instead of an AMI ID The following example uses a launch template to specify the configuration for the instances in the fleet. In this example, for `ImageId`, instead of specifying an AMI ID, the AMI is referenced with a System Manager parameter. On instance launch, the Systems Manager parameter will resolve to an AMI ID. In this example, the Systems Manager parameter is specified in a valid format: `resolve:ssm:golden-ami`. There are other valid formats for the Systems Manager parameter. For more information, see [Use a Systems Manager parameter instead of an AMI ID](create-launch-template.md#use-an-ssm-parameter-instead-of-an-ami-id). **Note** The fleet type must be of type `instant`. Other fleet types do not support specifying a System Manager parameter instead of an AMI ID. ``` { "LaunchTemplateData": { "ImageId": "resolve:ssm:golden-ami", "InstanceType": "m5.4xlarge", "TagSpecifications": [{ "ResourceType": "instance", "Tags": [{ "Key": "Name", "Value": "webserver" }] }] } } ``` # Set a spending limit for your EC2 Fleet or Spot Fleet You can set a limit on how much you're willing to spend per hour on your EC2 Fleet or Spot Fleet. When your spending limit is reached, the fleet stops launching instances, even if the target capacity hasn't been reached. There are separate spending limits for On-Demand Instances and Spot Instances. **To configure a spending limit for On-Demand Instances and Spot Instances in your EC2 Fleet** Use the [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command and the following parameters: + For On-Demand Instances: In the `OnDemandOptions` structure, specify your spending limit in the `MaxTotalPrice` field. + For Spot Instances: In the `SpotOptions` structure, specify your spending limit in the `MaxTotalPrice` field. **To configure a spending limit for On-Demand Instances and Spot Instances in your Spot Fleet** You can use the Amazon EC2 console or the AWS CLI to configure your spending limit. (Console) When creating the Spot Fleet, select the **Set maximum cost for Spot Instances** checkbox, and then enter a value for **Set your max cost (per hour)**. For more information, see step 6.e. in [Create a Spot Fleet request using defined parameters](create-spot-fleet.md#create-spot-fleet-advanced). (AWS CLI) Use the [request-spot-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/request-spot-fleet.html) command and the following parameters: + For On-Demand Instances: Specify your spending limit in the `OnDemandMaxTotalPrice` field. + For Spot Instances: Specify your spending limit in the `SpotMaxTotalPrice` field. ## Examples The following examples show two different scenarios. In the first example, the fleet stops launching On-Demand Instances when it has met the target capacity set for On-Demand Instances (`OnDemandTargetCapacity`). In the second example, the fleet stops launching On-Demand Instances when it has reached the maximum amount you’re willing to pay per hour for On-Demand Instances (`MaxTotalPrice`). **Example: Stop launching On-Demand Instances when target capacity is reached** Given a request for `m4.large` On-Demand Instances, where: + On-Demand Price: \$10.10 per hour + `OnDemandTargetCapacity`: 10 + `MaxTotalPrice`: \$11.50 The fleet launches 10 On-Demand Instances because the total of \$11.00 (10 instances x \$10.10) does not exceed the `MaxTotalPrice` of \$11.50 for On-Demand Instances. **Example: Stop launching On-Demand Instances when maximum total price is reached** Given a request for `m4.large` On-Demand Instances, where: + On-Demand Price: \$10.10 per hour + `OnDemandTargetCapacity`: 10 + `MaxTotalPrice`: \$10.80 If the fleet launches the On-Demand target capacity (10 On-Demand Instances), the total cost per hour would be \$11.00. This is more than the amount (\$10.80) specified for `MaxTotalPrice` for On-Demand Instances. To prevent spending more than you're willing to pay, the fleet launches only 8 On-Demand Instances (below the On-Demand target capacity) because launching more would exceed the `MaxTotalPrice` for On-Demand Instances. ## Burstable performance instances If you launch your Spot Instances using a [burstable performance instance type](burstable-performance-instances.md), and if you plan to use your burstable performance Spot Instances immediately and for a short duration, with no idle time for accruing CPU credits, we recommend that you launch them in [Standard mode](burstable-performance-instances-standard-mode.md) to avoid paying higher costs. If you launch burstable performance Spot Instances in [Unlimited mode](burstable-performance-instances-unlimited-mode.md) and burst CPU immediately, you'll spend surplus credits for bursting. If you use the instance for a short duration, the instance doesn't have time to accrue CPU credits to pay down the surplus credits, and you are charged for the surplus credits when you terminate the instance. Unlimited mode is suitable for burstable performance Spot Instances only if the instance runs long enough to accrue CPU credits for bursting. Otherwise, paying for surplus credits makes burstable performance Spot Instances more expensive than using other instances. For more information, see [When to use unlimited mode versus fixed CPU](burstable-performance-instances-unlimited-mode-concepts.md#when-to-use-unlimited-mode). Launch credits are meant to provide a productive initial launch experience for T2 instances by providing sufficient compute resources to configure the instance. Repeated launches of T2 instances to access new launch credits is not permitted. If you require sustained CPU, you can earn credits (by idling over some period), use [Unlimited mode](burstable-performance-instances-unlimited-mode.md) for T2 Spot Instances, or use an instance type with dedicated CPU. # Specify attributes for instance type selection for EC2 Fleet or Spot Fleet When you create an EC2 Fleet or Spot Fleet, you must specify one or more instance types for configuring the On-Demand Instances and Spot Instances in the fleet. As an alternative to manually specifying the instance types, you can specify the attributes that an instance must have, and Amazon EC2 will identify all the instance types with those attributes. This is known as *attribute-based instance type selection*. For example, you can specify the minimum and maximum number of vCPUs required for your instances, and the fleet will launch the instances using any available instance types that meet those vCPU requirements. Attribute-based instance type selection is ideal for workloads and frameworks that can be flexible about what instance types they use, such as when running containers or web fleets, processing big data, and implementing continuous integration and deployment (CI/CD) tooling. **Benefits** Attribute-based instance type selection has the following benefits: + **Easily use the right instance types** – With so many instance types available, finding the right instance types for your workload can be time consuming. When you specify instance attributes, the instance types will automatically have the required attributes for your workload. + **Simplified configuration** – To manually specify multiple instance types for a fleet, you must create a separate launch template override for each instance type. But with attribute-based instance type selection, to provide multiple instance types, you need only specify the instance attributes in the launch template or in a launch template override. + **Automatic use of new instance types** – When you specify instance attributes rather than instance types, your fleet can use newer generation instance types as they’re released, "future proofing" the fleet's configuration. + **Instance type flexibility** – When you specify instance attributes rather than instance types, the fleet can select from a wide range of instance types for launching Spot Instances, which adheres to the [Spot best practice of instance type flexibility](spot-best-practices.md#be-instance-type-flexible). **Topics** + [ ## How attribute-based instance type selection works ](#ec2fleet-abs-how-it-works) + [ ## Price protection ](#ec2fleet-abs-price-protection) + [ ## Performance protection ](#ec2fleet-abis-performance-protection) + [ ## Considerations ](#ec2fleet-abs-considerations) + [ ## Create an EC2 Fleet with attribute-based instance type selection ](#abs-create-ec2-fleet) + [ ## Create a Spot Fleet with attribute-based instance type selection ](#abs-create-spot-fleet) + [ ## Examples of EC2 Fleet configurations that are valid and not valid ](#ec2fleet-abs-example-configs) + [ ## Examples of Spot Fleet configurations that are valid and not valid ](#spotfleet-abs-example-configs) + [ ## Preview instance types with specified attributes ](#ec2fleet-get-instance-types-from-instance-requirements) ## How attribute-based instance type selection works To use attribute-based instance type selection in your fleet configuration, you replace the list of instance types with a list of instance attributes that your instances require. EC2 Fleet or Spot Fleet will launch instances on any available instance types that have the specified instance attributes. **Topics** + [ ### Types of instance attributes ](#ef-abs-instance-attribute-types) + [ ### Where to configure attribute-based instance type selection ](#ef-abs-where-to-configure) + [ ### How EC2 Fleet or Spot Fleet uses attribute-based instance type selection when provisioning a fleet ](#how-ef-uses-abs) ### Types of instance attributes There are several instance attributes that you can specify to express your compute requirements, such as: + **vCPU count** – The minimum and maximum number of vCPUs per instance. + **Memory** – The minimum and maximum GiBs of memory per instance. + **Local storage** – Whether to use EBS or instance store volumes for local storage. + **Burstable performance** – Whether to use the T instance family, including T4g, T3a, T3, and T2 types. For a description of each attribute and the default values, see [InstanceRequirements](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_InstanceRequirements.html) in the *Amazon EC2 API Reference*. ### Where to configure attribute-based instance type selection Depending on whether you use the console or the AWS CLI, you can specify the instance attributes for attribute-based instance type selection as follows: In the console, you can specify the instance attributes in the following fleet configuration components: + In a launch template, and then reference the launch template in the fleet request + (Spot Fleet only) In the fleet request In the AWS CLI, you can specify the instance attributes in one or all of the following fleet configuration components: + In a launch template, and then reference the launch template in the fleet request + In a launch template override If you want a mix of instances that use different AMIs, you can specify instance attributes in multiple launch template overrides. For example, different instance types can use x86 and Arm-based processors. + (Spot Fleet only) In a launch specification ### How EC2 Fleet or Spot Fleet uses attribute-based instance type selection when provisioning a fleet EC2 Fleet or Spot Fleet provisions a fleet in the following way: + It identifies the instance types that have the specified attributes. + It uses price protection to determine which instance types to exclude. + It determines the capacity pools from which it will consider launching the instances based on the AWS Regions or Availability Zones that have matching instance types. + It applies the specified allocation strategy to determine from which capacity pools to launch the instances. Note that attribute-based instance type selection does not pick the capacity pools from which to provision the fleet; that's the job of the [allocation strategies](ec2-fleet-allocation-strategy.md). If you specify an allocation strategy, the fleet will launch instances according to the specified allocation strategy. + For Spot Instances, attribute-based instance type selection supports the **price capacity optimized**, **capacity optimized**, and **lowest price** allocation strategies. Note that we don't recommend the **lowest price** Spot allocation strategy because it has the highest risk of interruption for your Spot Instances. + For On-Demand Instances, attribute-based instance type selection supports the **lowest price** allocation strategy. + If there is no capacity for the instance types with the specified instance attributes, no instances can be launched, and the fleet returns an error. ## Price protection Price protection is a feature that prevents your EC2 Fleet or Spot Fleet from using instance types that you would consider too expensive even if they happen to fit the attributes that you specified. To use price protection, you set a price threshold. Then, when Amazon EC2 selects instance types with your attributes, it excludes instance types priced above your threshold. The way that Amazon EC2 calculates the price threshold is as follows: + Amazon EC2 first identifies the lowest priced instance type from those that match your attributes. + Amazon EC2 then takes the value (expressed as a percentage) that you specified for the price protection parameter and multiplies it with the price of the identified instance type. The result is the price that is used as the price threshold. There are separate price thresholds for On-Demand Instances and Spot Instances. When you create a fleet with attribute-based instance type selection, price protection is enabled by default. You can keep the default values, or you can specify your own. You can also turn off price protection. To indicate no price protection threshold, specify a high percentage value, such as `999999`. **Topics** + [ ### How the lowest priced instance type is identified ](#ec2fleet-abs-price-protection-lowest-priced) + [ ### On-Demand Instance price protection ](#ec2fleet-abs-on-demand-price-protection) + [ ### Spot Instance price protection ](#ec2fleet-abs-spot-price-protection) + [ ### Specify the price protection threshold ](#ec2fleet-abs-specify-price-protection) ### How the lowest priced instance type is identified Amazon EC2 determines the price to base the price threshold on by identifying the instance type with the lowest price from those that match your specified attributes. It does this in the following way: + It first looks at the current generation C, M, or R instance types that match your attributes. If it finds any matches, it identifies the lowest priced instance type. + If there is no match, it then looks at any current generation instance types that match your attributes. If it finds any matches, it identifies the lowest priced instance type. + If there is no match, it then looks at any previous generation instance types that match your attributes, and identifies the lowest priced instance type. ### On-Demand Instance price protection The price protection threshold for On-Demand instance types is calculated *as a percentage higher* than the identified lowest priced On-Demand instance type (`OnDemandMaxPricePercentageOverLowestPrice`). You specify the percentage higher that you're willing to pay. If you don't specify this parameter, then a default value of `20` is used to calculate a price protection threshold of 20% higher than the identified price. For example, if the identified On-Demand instance price is `0.4271`, and you specify `25`, then the price threshold is 25% more than `0.4271`. It is calculated as follows: `0.4271 * 1.25 = 0.533875`. The calculated price is the maximum you're willing to pay for On-Demand Instances, and, in this example, Amazon EC2 will exclude any On-Demand instance types that cost more than `0.533875`. ### Spot Instance price protection By default, Amazon EC2 will automatically apply optimal Spot Instance price protection to consistently select from a wide range of instance types. You can also manually set the price protection yourself. However, letting Amazon EC2 do it for you can improve the likelihood that your Spot capacity is fulfilled. You can manually specify the price protection using one of the following options. If you manually set the price protection, we recommend using the first option. + **A *percentage of* the identified lowest priced *On-Demand* instance type** [`MaxSpotPriceAsPercentageOfOptimalOnDemandPrice`] For example, if the identified On-Demand instance type price is `0.4271`, and you specify `60`, then the price threshold is 60% of `0.4271`. It is calculated as follows: `0.4271 * 0.60 = 0.25626`. The calculated price is the maximum you're willing to pay for Spot Instances, and, in this example, Amazon EC2 will exclude any Spot instance types that cost more than `0.25626`. + **A *percentage higher than* the identified lowest priced *Spot* instance type** [`SpotMaxPricePercentageOverLowestPrice`] For example, if the identified Spot instance type price is `0.1808`, and you specify `25`, then the price threshold is 25% more than `0.1808`. It is calculated as follows: `0.1808 * 1.25 = 0.226`. The calculated price is the maximum you're willing to pay for Spot Instances, and, in this example, Amazon EC2 will exclude any Spot instance types that cost more than `0.266`. We do not recommend using this parameter because Spot prices can fluctuate, and therefore your price protection threshold might also fluctuate. ### Specify the price protection threshold **To specify the price protection threshold using the AWS CLI** While creating an EC2 Fleet or Spot Fleet using the AWS CLI, configure the fleet for attribute-based instance type selection, and then do the following: + To specify the On-Demand Instance price protection threshold, in the JSON configuration file, in the `InstanceRequirements` structure, for `OnDemandMaxPricePercentageOverLowestPrice`, enter the price protection threshold as a percentage. + To specify the Spot Instance price protection threshold, in the JSON configuration file, in the `InstanceRequirements` structure, specify *one* of the following parameters: + For `MaxSpotPriceAsPercentageOfOptimalOnDemandPrice`, enter the price protection threshold as a percentage. + For `SpotMaxPricePercentageOverLowestPrice`, enter the price protection threshold as a percentage. For more information, see [Create an EC2 Fleet with attribute-based instance type selection](#abs-create-ec2-fleet) or [Create a Spot Fleet with attribute-based instance type selection](#abs-create-spot-fleet). **(Spot Fleet only) To specify the price protection threshold using the console** While creating a Spot Fleet in the console, configure the fleet for attribute-based instance type selection, and then do the following: + To specify the On-Demand Instance price protection threshold, under **Additional instance attribute**, choose **On-demand price protection**, choose **Add attribute**, and then enter the price protection threshold as a percentage. + To specify the Spot Instance price protection threshold, **Additional instance attribute**, choose **Spot price protection**, choose **Add attribute**, choose a base value on which to base your price, and then enter the price protection threshold as a percentage. **Note** When creating the fleet, if you set `TargetCapacityUnitType` to `vcpu` or `memory-mib`, the price protection threshold is applied based on the per-vCPU or per-memory price instead of the per-instance price. ## Performance protection *Performance protection* is a feature that ensures your EC2 Fleet or Spot Fleet uses instance types that are similar to or exceed a specified performance baseline. To use performance protection, you specify an instance family as a baseline reference. The capabilities of the specified instance family establish the lowest acceptable level of performance. When Amazon EC2 selects instance types for your fleet, it considers your specified attributes and the performance baseline. Instance types that fall below the performance baseline are automatically excluded from selection, even if they match your other specified attributes. This ensures that all selected instance types offer performance similar to or better than the baseline established by the specified instance family. Amazon EC2 uses this baseline to guide instance type selection, but there is no guarantee that the selected instance types will always exceed the baseline for every application. Currently, this feature only supports CPU performance as a baseline performance factor. The CPU performance of the specified instance family's CPU processor serves as the performance baseline, ensuring that selected instance types are similar to or exceed this baseline. Instance families with the same CPU processors lead to the same filtering results, even if their network or disk performance differs. For example, specifying either `c6in` or `c6i` as the baseline reference would produce identical performance-based filtering results because both instance families use the same CPU processor. **Unsupported instance families** The following instance families are **not** supported for performance protection: + **General purpose:** Mac1 \$1 Mac2 \$1 Mac2-m1ultra \$1 Mac2-m2 \$1 Mac2-m2pro \$1 M1 \$1 M2 \$1 T1 + **Compute optimized:** C1 + **Memory optimized:** U-3tb1 \$1 U-6tb1 \$1 U-9tb1 \$1 U-12tb1 \$1 U-18tb1 \$1 u-24tb1 \$1 U7i-12tb \$1 U7in-16tb \$1 U7in-24tb \$1 U7in-32tb + **Accelerated computing:** G3 \$1 G3s \$1 P3dn \$1 P4d \$1 P5 + **High-performance computing:** Hpc7g If you enable performance protection by specifying a supported instance family, the returned instance types will exclude the above unsupported instance families. If you specify an unsupported instance family as a value for baseline performance, the API returns an empty response for [GetInstanceTypesFromInstanceRequirements](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_GetInstanceTypesFromInstanceRequirements.html) and an exception for [CreateFleet](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_CreateFleet.html), [RequestSpotFleet](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_RequestSpotFleet.html), [ModifyFleet](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_ModifyFleet.html), and [ModifySpotFleetRequest](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_ModifySpotFleetRequest.html). **Example: Set a CPU performance baseline** In the following example, the instance requirement is to launch with instance types that have CPU cores that are as performant as the `c6i` instance family. This will filter out instance types with less performant CPU processors, even if they meet your other specified instance requirements such as the number of vCPUs. For example, if your specified instance attributes include 4 vCPUs and 16 GB of memory, an instance type with these attributes but with lower CPU performance than `c6i` will be excluded from selection. ``` "BaselinePerformanceFactors": { "Cpu": { "References": [ { "InstanceFamily": "c6i" } ] } ``` ## Considerations + You can specify either instance types or instance attributes in an EC2 Fleet or Spot Fleet, but not both at the same time. When using the CLI, the launch template overrides will override the launch template. For example, if the launch template contains an instance type and the launch template override contains instance attributes, the instances that are identified by the instance attributes will override the instance type in the launch template. + When using the CLI, when you specify instance attributes as overrides, you can't also specify weights or priorities. + You can specify a maximum of four `InstanceRequirements` structures in a request configuration. ## Create an EC2 Fleet with attribute-based instance type selection You can configure an EC2 Fleet to use attribute-based instance type selection. It is not possible to create an EC2 Fleet using the Amazon EC2 console. The attributes for attribute-based instance type selection are specified in the `InstanceRequirements` structure. When `InstanceRequirements` is included in the fleet configuration, `InstanceType` and `WeightedCapacity` must be excluded; they can't determine the fleet configuration at the same time as instance attributes. In the AWS CLI and PowerShell examples, the following attributes are specified: + `VCpuCount` – A minimum of 2 vCPUs and a maximum of 4 vCPUs. If you don't need a maximum limit, you can omit the maximum value. + `MemoryMiB` – A minimum of 8 GiB of memory and a maximum of 16 GiB. If you don't need a maximum limit, you can omit the maximum value. This configuration identifies any instance types with 2 to 4 vCPUs and 8 to 16 GiB of memory. However, price protection and the allocation strategy might exclude some instance types when [EC2 Fleet provisions the fleet](#how-ef-uses-abs). ------ #### [ AWS CLI ] **To create an EC2 Fleet with attribute-based instance type selection** Use the [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command to create an EC2 Fleet. Specify the fleet configuration in a JSON file. ``` aws ec2 create-fleet \ --region us-east-1 \ --cli-input-json file://file_name.json ``` The following example `file_name.json` file contains the parameters that configure an EC2 Fleet to use attribute-based instance type selection. ``` { "SpotOptions": { "AllocationStrategy": "price-capacity-optimized" }, "LaunchTemplateConfigs": [{ "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [{ "InstanceRequirements": { "VCpuCount": { "Min": 2, "Max": 4 }, "MemoryMiB": { "Min": 8192, "Max": 16384 } } }] }], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ------ #### [ PowerShell ] **To create an EC2 Fleet with attribute-based instance type selection** Use the [New-EC2Fleet](https://docs.aws.amazon.com/powershell/latest/reference/items/New-EC2Fleet.html) cmdlet. ``` $vcpuCount = New-Object Amazon.EC2.Model.VCpuCountRangeRequest $vcpuCount.Min = 2 $vcpuCount.Max = 4 $memoryMiB = New-Object Amazon.EC2.Model.MemoryMiBRequest $memoryMiB.Min = 8192 $memoryMiB.Max = 16384 $instanceRequirements = New-Object Amazon.EC2.Model.InstanceRequirementsRequest $instanceRequirements.VCpuCount = $vcpuCount $instanceRequirements.MemoryMiB = $memoryMiB $launchTemplateSpec = New-Object Amazon.EC2.Model.FleetLaunchTemplateSpecificationRequest $launchTemplateSpec.LaunchTemplateName = "my-launch-template" $launchTemplateSpec.Version = "1" $override = New-Object Amazon.EC2.Model.FleetLaunchTemplateOverridesRequest $override.InstanceRequirements = $instanceRequirements $launchTemplateConfig = New-Object Amazon.EC2.Model.FleetLaunchTemplateConfigRequest $launchTemplateConfig.LaunchTemplateSpecification = $launchTemplateSpec $launchTemplateConfig.Overrides = @($override) New-EC2Fleet ` -SpotOptions_AllocationStrategy "price-capacity-optimized" ` -LaunchTemplateConfig @($launchTemplateConfig) ` -TargetCapacitySpecification_DefaultTargetCapacityType "spot" ` -TargetCapacitySpecification_TotalTargetCapacity 20 ` -Type "instant" ``` ------ ## Create a Spot Fleet with attribute-based instance type selection You can configure a fleet to use attribute-based instance type selection. The attributes for attribute-based instance type selection are specified in the `InstanceRequirements` structure. When `InstanceRequirements` is included in the fleet configuration, `InstanceType` and `WeightedCapacity` must be excluded; they can't determine the fleet configuration at the same time as instance attributes. In the AWS CLI and PowerShell examples, the following attributes are specified: + `VCpuCount` – A minimum of 2 vCPUs and a maximum of 4 vCPUs. If you don't need a maximum limit, you can omit the maximum value. + `MemoryMiB` – A minimum of 8 GiB of memory and a maximum of 16 GiB. If you don't need a maximum limit, you can omit the maximum value. This configuration identifies any instance types that have 2 to 4 vCPUs and 8 to 16 GiB of memory. However, price protection and the allocation strategy might exclude some instance types when [Spot Fleet provisions the fleet](#how-ef-uses-abs). ------ #### [ Console ] **To configure a Spot Fleet for attribute-based instance type selection** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**, and then choose **Request Spot Instances**. 1. Follow the steps to create a Spot Fleet. For more information, see [Create a Spot Fleet request using defined parameters](create-spot-fleet.md#create-spot-fleet-advanced). While creating the Spot Fleet, configure the fleet for attribute-based instance type selection as follows: 1. For **Instance type requirements**, choose **Specify instance attributes that match your compute requirements**. 1. For **vCPUs**, enter the desired minimum and maximum number of vCPUs. To specify no limit, select **No minimum**, **No maximum**, or both. 1. For **Memory (GiB)**, enter the desired minimum and maximum amount of memory. To specify no limit, select **No minimum**, **No maximum**, or both. 1. (Optional) For **Additional instance attributes**, you can optionally specify one or more attributes to express your compute requirements in more detail. Each additional attribute adds further constraints to your request. 1. (Optional) Expand **Preview matching instance types** to view the instance types that have your specified attributes. ------ #### [ AWS CLI ] **To configure a Spot Fleet for attribute-based instance type selection** Use the [request-spot-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/request-spot-fleet.html) command to create a Spot Fleet. Specify the fleet configuration in a JSON file. ``` aws ec2 request-spot-fleet \ --region us-east-1 \ --spot-fleet-request-config file://file_name.json ``` The following example `file_name.json` file contains the parameters that configure a Spot Fleet to use attribute-based instance type selection. ``` { "AllocationStrategy": "priceCapacityOptimized", "TargetCapacity": 20, "Type": "request", "LaunchTemplateConfigs": [{ "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [{ "InstanceRequirements": { "VCpuCount": { "Min": 2, "Max": 4 }, "MemoryMiB": { "Min": 8192, "Max": 16384 } } }] }] } ``` ------ #### [ PowerShell ] **To configure a Spot Fleet for attribute-based instance type selection** Use the [Request-EC2SpotFleet](https://docs.aws.amazon.com/powershell/latest/reference/items/Request-EC2SpotFleet.html) cmdlet. ``` $vcpuCount = New-Object Amazon.EC2.Model.VCpuCountRange $vcpuCount.Min = 2 $vcpuCount.Max = 4 $memoryMiB = New-Object Amazon.EC2.Model.MemoryMiB $memoryMiB.Min = 8192 $memoryMiB.Max = 16384 $instanceRequirements = New-Object Amazon.EC2.Model.InstanceRequirements $instanceRequirements.VCpuCount = $vcpuCount $instanceRequirements.MemoryMiB = $memoryMiB $launchTemplateSpec = New-Object Amazon.EC2.Model.FleetLaunchTemplateSpecification $launchTemplateSpec.LaunchTemplateName = "my-launch-template" $launchTemplateSpec.Version = "1" $override = New-Object Amazon.EC2.Model.LaunchTemplateOverrides $override.InstanceRequirements = $instanceRequirements $launchTemplateConfig = New-Object Amazon.EC2.Model.LaunchTemplateConfig $launchTemplateConfig.LaunchTemplateSpecification = $launchTemplateSpec $launchTemplateConfig.Overrides = @($override) Request-EC2SpotFleet ` -SpotFleetRequestConfig_AllocationStrategy "PriceCapacityOptimized" ` -SpotFleetRequestConfig_TargetCapacity 20 ` -SpotFleetRequestConfig_Type "Request" ` -SpotFleetRequestConfig_LaunchTemplateConfig $launchTemplateConfig ``` ------ ## Examples of EC2 Fleet configurations that are valid and not valid If you use the AWS CLI to create an EC2 Fleet, you must make sure that your fleet configuration is valid. The following examples show configurations that are valid and not valid. Configurations are considered not valid when they contain the following: + A single `Overrides` structure with both `InstanceRequirements` and `InstanceType` + Two `Overrides` structures, one with `InstanceRequirements` and the other with `InstanceType` + Two `InstanceRequirements` structures with overlapping attribute values within the same `LaunchTemplateSpecification` **Topics** + [ ### Valid configuration: Single launch template with overrides ](#ef-abs-example-config1) + [ ### Valid configuration: Single launch template with multiple InstanceRequirements ](#ef-abs-example-config2) + [ ### Valid configuration: Two launch templates, each with overrides ](#ef-abs-example-config3) + [ ### Valid configuration: Only `InstanceRequirements` specified, no overlapping attribute values ](#ef-abs-example-config4) + [ ### Configuration not valid: `Overrides` contain `InstanceRequirements` and `InstanceType` ](#ef-abs-example-config5) + [ ### Configuration not valid: Two `Overrides` contain `InstanceRequirements` and `InstanceType` ](#ef-abs-example-config6) + [ ### Configuration not valid: Overlapping attribute values ](#ef-abs-example-config7) ### Valid configuration: Single launch template with overrides The following configuration is valid. It contains one launch template and one `Overrides` structure containing one `InstanceRequirements` structure. A text explanation of the example configuration follows. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "My-launch-template", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 2, "Max": 8 }, "MemoryMib": { "Min": 0, "Max": 10240 }, "MemoryGiBPerVCpu": { "Max": 10000 }, "RequireHibernateSupport": true } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 5000, "DefaultTargetCapacityType": "spot", "TargetCapacityUnitType": "vcpu" } } } ``` ****`InstanceRequirements`**** To use attribute-based instance selection, you must include the `InstanceRequirements` structure in your fleet configuration, and specify the desired attributes for the instances in the fleet. In the preceding example, the following instance attributes are specified: + `VCpuCount` – The instance types must have a minimum of 2 and a maximum of 8 vCPUs. + `MemoryMiB` – The instance types must have a maximum of 10240 MiB of memory. A minimum of 0 indicates no minimum limit. + `MemoryGiBPerVCpu` – The instance types must have a maximum of 10,000 GiB of memory per vCPU. The `Min` parameter is optional. By omitting it, you indicate no minimum limit. **`TargetCapacityUnitType`** The `TargetCapacityUnitType` parameter specifies the unit for the target capacity. In the example, the target capacity is `5000` and the target capacity unit type is `vcpu`, which together specify a desired target capacity of 5,000 vCPUs. EC2 Fleet will launch enough instances so that the total number of vCPUs in the fleet is 5,000 vCPUs. ### Valid configuration: Single launch template with multiple InstanceRequirements The following configuration is valid. It contains one launch template and one `Overrides` structure containing two `InstanceRequirements` structures. The attributes specified in `InstanceRequirements` are valid because the values do not overlap—the first `InstanceRequirements` structure specifies a `VCpuCount` of 0-2 vCPUs, while the second `InstanceRequirements` structure specifies 4-8 vCPUs. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } }, { "InstanceRequirements": { "VCpuCount": { "Min": 4, "Max": 8 }, "MemoryMiB": { "Min": 0 } } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 1, "DefaultTargetCapacityType": "spot" } } } ``` ### Valid configuration: Two launch templates, each with overrides The following configuration is valid. It contains two launch templates, each with one `Overrides` structure containing one `InstanceRequirements` structure. This configuration is useful for `arm` and `x86` architecture support in the same fleet. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "armLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } }, { "LaunchTemplateSpecification": { "LaunchTemplateName": "x86LaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 1, "DefaultTargetCapacityType": "spot" } } } ``` ### Valid configuration: Only `InstanceRequirements` specified, no overlapping attribute values The following configuration is valid. It contains two `LaunchTemplateSpecification` structures, each with a launch template and an `Overrides` structure containing an `InstanceRequirements` structure. The attributes specified in `InstanceRequirements` are valid because the values do not overlap—the first `InstanceRequirements` structure specifies a `VCpuCount` of 0-2 vCPUs, while the second `InstanceRequirements` structure specifies 4-8 vCPUs. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } } ] }, { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyOtherLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 4, "Max": 8 }, "MemoryMiB": { "Min": 0 } } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 1, "DefaultTargetCapacityType": "spot" } } } ``` ### Configuration not valid: `Overrides` contain `InstanceRequirements` and `InstanceType` The following configuration is not valid. The `Overrides` structure contains both `InstanceRequirements` and `InstanceType`. For the `Overrides`, you can specify either `InstanceRequirements` or `InstanceType`, but not both. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } }, { "InstanceType": "m5.large" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 1, "DefaultTargetCapacityType": "spot" } } } ``` ### Configuration not valid: Two `Overrides` contain `InstanceRequirements` and `InstanceType` The following configuration is not valid. The `Overrides` structures contain both `InstanceRequirements` and `InstanceType`. You can specify either `InstanceRequirements` or `InstanceType`, but not both, even if they're in different `Overrides` structures. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } } ] }, { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyOtherLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceType": "m5.large" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 1, "DefaultTargetCapacityType": "spot" } } } ``` ### Configuration not valid: Overlapping attribute values The following configuration is not valid. The two `InstanceRequirements` structures each contain `"VCpuCount": {"Min": 0, "Max": 2}`. The values for these attributes overlap, which will result in duplicate capacity pools. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } }, { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 1, "DefaultTargetCapacityType": "spot" } } } ``` ## Examples of Spot Fleet configurations that are valid and not valid If you use the AWS CLI to create a Spot Fleet, you must make sure that your fleet configuration is valid. The following examples show configurations that are valid and not valid. Configurations are considered not valid when they contain the following: + A single `Overrides` structure with both `InstanceRequirements` and `InstanceType` + Two `Overrides` structures, one with `InstanceRequirements` and the other with `InstanceType` + Two `InstanceRequirements` structures with overlapping attribute values within the same `LaunchTemplateSpecification` **Topics** + [ ### Valid configuration: Single launch template with overrides ](#sf-abs-example-config1) + [ ### Valid configuration: Single launch template with multiple InstanceRequirements ](#sf-abs-example-config2) + [ ### Valid configuration: Two launch templates, each with overrides ](#sf-abs-example-config3) + [ ### Valid configuration: Only `InstanceRequirements` specified, no overlapping attribute values ](#sf-abs-example-config4) + [ ### Configuration not valid: `Overrides` contain `InstanceRequirements` and `InstanceType` ](#sf-abs-example-config5) + [ ### Configuration not valid: Two `Overrides` contain `InstanceRequirements` and `InstanceType` ](#sf-abs-example-config6) + [ ### Configuration not valid: Overlapping attribute values ](#sf-abs-example-config7) ### Valid configuration: Single launch template with overrides The following configuration is valid. It contains one launch template and one `Overrides` structure containing one `InstanceRequirements` structure. A text explanation of the example configuration follows. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "My-launch-template", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 2, "Max": 8 }, "MemoryMib": { "Min": 0, "Max": 10240 }, "MemoryGiBPerVCpu": { "Max": 10000 }, "RequireHibernateSupport": true } } ] } ], "TargetCapacity": 5000, "OnDemandTargetCapacity": 0, "TargetCapacityUnitType": "vcpu" } } ``` ****`InstanceRequirements`**** To use attribute-based instance selection, you must include the `InstanceRequirements` structure in your fleet configuration, and specify the desired attributes for the instances in the fleet. In the preceding example, the following instance attributes are specified: + `VCpuCount` – The instance types must have a minimum of 2 and a maximum of 8 vCPUs. + `MemoryMiB` – The instance types must have a maximum of 10240 MiB of memory. A minimum of 0 indicates no minimum limit. + `MemoryGiBPerVCpu` – The instance types must have a maximum of 10,000 GiB of memory per vCPU. The `Min` parameter is optional. By omitting it, you indicate no minimum limit. **`TargetCapacityUnitType`** The `TargetCapacityUnitType` parameter specifies the unit for the target capacity. In the example, the target capacity is `5000` and the target capacity unit type is `vcpu`, which together specify a desired target capacity of 5,000 vCPUs. Spot Fleet will launch enough instances so that the total number of vCPUs in the fleet is 5,000 vCPUs. ### Valid configuration: Single launch template with multiple InstanceRequirements The following configuration is valid. It contains one launch template and one `Overrides` structure containing two `InstanceRequirements` structures. The attributes specified in `InstanceRequirements` are valid because the values do not overlap—the first `InstanceRequirements` structure specifies a `VCpuCount` of 0-2 vCPUs, while the second `InstanceRequirements` structure specifies 4-8 vCPUs. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } }, { "InstanceRequirements": { "VCpuCount": { "Min": 4, "Max": 8 }, "MemoryMiB": { "Min": 0 } } } ] } ], "TargetCapacity": 1, "OnDemandTargetCapacity": 0, "Type": "maintain" } } ``` ### Valid configuration: Two launch templates, each with overrides The following configuration is valid. It contains two launch templates, each with one `Overrides` structure containing one `InstanceRequirements` structure. This configuration is useful for `arm` and `x86` architecture support in the same fleet. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "armLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } }, { "LaunchTemplateSpecification": { "LaunchTemplateName": "x86LaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } } ] } ], "TargetCapacity": 1, "OnDemandTargetCapacity": 0, "Type": "maintain" } } ``` ### Valid configuration: Only `InstanceRequirements` specified, no overlapping attribute values The following configuration is valid. It contains two `LaunchTemplateSpecification` structures, each with a launch template and an `Overrides` structure containing an `InstanceRequirements` structure. The attributes specified in `InstanceRequirements` are valid because the values do not overlap—the first `InstanceRequirements` structure specifies a `VCpuCount` of 0-2 vCPUs, while the second `InstanceRequirements` structure specifies 4-8 vCPUs. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } } ] }, { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyOtherLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 4, "Max": 8 }, "MemoryMiB": { "Min": 0 } } } ] } ], "TargetCapacity": 1, "OnDemandTargetCapacity": 0, "Type": "maintain" } } ``` ### Configuration not valid: `Overrides` contain `InstanceRequirements` and `InstanceType` The following configuration is not valid. The `Overrides` structure contains both `InstanceRequirements` and `InstanceType`. For the `Overrides`, you can specify either `InstanceRequirements` or `InstanceType`, but not both. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } }, { "InstanceType": "m5.large" } ] } ], "TargetCapacity": 1, "OnDemandTargetCapacity": 0, "Type": "maintain" } } ``` ### Configuration not valid: Two `Overrides` contain `InstanceRequirements` and `InstanceType` The following configuration is not valid. The `Overrides` structures contain both `InstanceRequirements` and `InstanceType`. You can specify either `InstanceRequirements` or `InstanceType`, but not both, even if they're in different `Overrides` structures. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } } ] }, { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyOtherLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceType": "m5.large" } ] } ], "TargetCapacity": 1, "OnDemandTargetCapacity": 0, "Type": "maintain" } } ``` ### Configuration not valid: Overlapping attribute values The following configuration is not valid. The two `InstanceRequirements` structures each contain `"VCpuCount": {"Min": 0, "Max": 2}`. The values for these attributes overlap, which will result in duplicate capacity pools. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "MyLaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } }, { "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 2 }, "MemoryMiB": { "Min": 0 } } } } ] } ], "TargetCapacity": 1, "OnDemandTargetCapacity": 0, "Type": "maintain" } } ``` ## Preview instance types with specified attributes You can use the [get-instance-types-from-instance-requirements](https://docs.aws.amazon.com/cli/latest/reference/ec2/get-instance-types-from-instance-requirements.html) command to preview the instance types that match the attributes that you specify. This is especially useful for working out what attributes to specify in your request configuration without launching any instances. Note that the command does not consider available capacity. **To preview a list of instance types by specifying attributes using the AWS CLI** 1. (Optional) To generate all of the possible attributes that can be specified, use the [get-instance-types-from-instance-requirements](https://docs.aws.amazon.com/cli/latest/reference/ec2/get-instance-types-from-instance-requirements.html) command and the `--generate-cli-skeleton` parameter. You can optionally direct the output to a file to save it by using `input > attributes.json`. ``` aws ec2 get-instance-types-from-instance-requirements \ --region us-east-1 \ --generate-cli-skeleton input > attributes.json ``` Expected output ``` { "DryRun": true, "ArchitectureTypes": [ "i386" ], "VirtualizationTypes": [ "hvm" ], "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 0 }, "MemoryMiB": { "Min": 0, "Max": 0 }, "CpuManufacturers": [ "intel" ], "MemoryGiBPerVCpu": { "Min": 0.0, "Max": 0.0 }, "ExcludedInstanceTypes": [ "" ], "InstanceGenerations": [ "current" ], "SpotMaxPricePercentageOverLowestPrice": 0, "OnDemandMaxPricePercentageOverLowestPrice": 0, "BareMetal": "included", "BurstablePerformance": "included", "RequireHibernateSupport": true, "NetworkInterfaceCount": { "Min": 0, "Max": 0 }, "LocalStorage": "included", "LocalStorageTypes": [ "hdd" ], "TotalLocalStorageGB": { "Min": 0.0, "Max": 0.0 }, "BaselineEbsBandwidthMbps": { "Min": 0, "Max": 0 }, "AcceleratorTypes": [ "gpu" ], "AcceleratorCount": { "Min": 0, "Max": 0 }, "AcceleratorManufacturers": [ "nvidia" ], "AcceleratorNames": [ "a100" ], "AcceleratorTotalMemoryMiB": { "Min": 0, "Max": 0 }, "NetworkBandwidthGbps": { "Min": 0.0, "Max": 0.0 }, "AllowedInstanceTypes": [ "" ] }, "MaxResults": 0, "NextToken": "" } ``` 1. Create a JSON configuration file using the output from the previous step, and configure it as follows: **Note** You must provide values for `ArchitectureTypes`, `VirtualizationTypes`, `VCpuCount`, and `MemoryMiB`. You can omit the other attributes; when omitted, the default values are used. For a description of each attribute and their default values, see [get-instance-types-from-instance-requirements](https://docs.aws.amazon.com/cli/latest/reference/ec2/get-instance-types-from-instance-requirements.html). 1. For `ArchitectureTypes`, specify one or more types of processor architecture. 1. For `VirtualizationTypes`, specify one or more types of virtualization. 1. For `VCpuCount`, specify the minimum and maximum number of vCPUs. To specify no minimum limit, for `Min`, specify `0`. To specify no maximum limit, omit the `Max` parameter. 1. For `MemoryMiB`, specify the minimum and maximum amount of memory in MiB. To specify no minimum limit, for `Min`, specify `0`. To specify no maximum limit, omit the `Max` parameter. 1. You can optionally specify one or more of the other attributes to further constrain the list of instance types that are returned. 1. To preview the instance types that have the attributes that you specified in the JSON file, use the [get-instance-types-from-instance-requirements](https://docs.aws.amazon.com/cli/latest/reference/ec2/get-instance-types-from-instance-requirements.html) command, and specify the name and path to your JSON file by using the `--cli-input-json` parameter. You can optionally format the output to appear in a table format. ``` aws ec2 get-instance-types-from-instance-requirements \ --cli-input-json file://attributes.json \ --output table ``` Example *attributes.json* file In this example, the required attributes are included in the JSON file. They are `ArchitectureTypes`, `VirtualizationTypes`, `VCpuCount`, and `MemoryMiB`. In addition, the optional `InstanceGenerations` attribute is also included. Note that for `MemoryMiB`, the `Max` value can be omitted to indicate that there is no limit. ``` { "ArchitectureTypes": [ "x86_64" ], "VirtualizationTypes": [ "hvm" ], "InstanceRequirements": { "VCpuCount": { "Min": 4, "Max": 6 }, "MemoryMiB": { "Min": 2048 }, "InstanceGenerations": [ "current" ] } } ``` Example output ``` ------------------------------------------ |GetInstanceTypesFromInstanceRequirements| +----------------------------------------+ || InstanceTypes || |+--------------------------------------+| || InstanceType || |+--------------------------------------+| || c4.xlarge || || c5.xlarge || || c5a.xlarge || || c5ad.xlarge || || c5d.xlarge || || c5n.xlarge || || d2.xlarge || ... ``` 1. After identifying instance types that meet your needs, make note of the instance attributes that you used so that you can use them when configuring your fleet request. # Use instance weighting to manage cost and performance of your EC2 Fleet or Spot Fleet With instance weighting, you assign a weight to each instance type in your EC2 Fleet or Spot Fleet to represent their compute capacity and performance relative to each other. Based on the weights, the fleet can use any combination of the specified instance types, as long as it can fulfil the desired target capacity. This can help you manage the cost and performance of your fleet. The weight represents the capacity units that an instance type contributes to the total target capacity. **Example: Use instance weighting for performance management** Suppose your fleet has two instance types, and you assign a different weight to each instance type to reflect how many you need of each to achieve the same performance, as follows: + `m5.large` – weight: 1 + `m5.2xlarge` – weight: 4 By assigning these weights, you're saying that you'd need 4 `m5.large` instances to achieve the same performance as 1 `m5.2xlarge`. To calculate how many instances of each instance type are needed for a given target capacity, use the following formula: `target capacity / weight = number of instances` If your target capacity is 8 units, the fleet could fulfill the target capacity with either `m5.large` or `m5.2xlarge`, or a mix of both, as follows: + 8 `m5.large` instances (capacity of 8 / weight of 1 = 8 instances) + 2 `m5.2xlarge` instances (capacity of 8 / weight of 4 = 2 instances) + 4 `m5.large` and 1 `m5.2xlarge` **Example: Use instance weighting for cost management** By default, the price that you specify is per *instance* hour. When you use the instance weighting feature, the price that you specify is per *unit* hour. You can calculate your price per unit hour by dividing your price for an instance type by the number of units that it represents. The fleet calculates the number of instances to launch by dividing the target capacity by the instance weight. If the result isn't an integer, the fleet rounds it up to the next integer, so that the size of your fleet is not below its target capacity. The fleet can select any pool that you specify in your launch specification, even if the capacity of the instances launched exceeds the requested target capacity. The following table includes examples of calculations to determine the price per unit for a fleet with a target capacity of 10. | Instance type | Instance weight | Target capacity | Number of instances launched | Price per instance hour | Price per unit hour | | --- | --- | --- | --- | --- | --- | | r3.xlarge | 2 | 10 | 5 (10 divided by 2) | \$10.05 | \$10.025 (.05 divided by 2) | | r3.8xlarge | 8 | 10 | 2 (10 divided by 8, result rounded up) | \$10.10 | \$10.0125 (.10 divided by 8) | Use the fleet instance weighting as follows to provision the target capacity that you want in the pools with the lowest price per unit at the time of fulfillment: 1. Set the target capacity for your fleet either in instances (the default) or in the units of your choice, such as vCPU, memory, storage, or throughput. 1. Set the price per unit. 1. For each launch specification, specify the weight, which is the number of units that the instance type represents toward the target capacity. **Instance weighting example** Consider a fleet request with the following configuration: + A target capacity of 24 + A launch specification with an instance type `r3.2xlarge` and a weight of 6 + A launch specification with an instance type `c3.xlarge` and a weight of 5 The weights represent the number of units that instance type represents toward the target capacity. If the first launch specification provides the lowest price per unit (price for `r3.2xlarge` per instance hour divided by 6), the fleet would launch four of these instances (24 divided by 6). If the second launch specification provides the lowest price per unit (price for `c3.xlarge` per instance hour divided by 5), the fleet would launch five of these instances (24 divided by 5, result rounded up). **Instance weighting and allocation strategy** Consider a fleet request with the following configuration: + A target capacity of 30 Spot Instances + A launch specification with an instance type `c3.2xlarge` and a weight of 8 + A launch specification with an instance type `m3.xlarge` and a weight of 8 + A launch specification with an instance type `r3.xlarge` and a weight of 8 The fleet would launch four instances (30 divided by 8, result rounded up). With the `diversified` strategy, the fleet launches one instance in each of the three pools, and the fourth instance in whichever of the three pools provides the lowest price per unit. # Use allocation strategies to determine how EC2 Fleet or Spot Fleet fulfills Spot and On-Demand capacity When you use multiple capacity pools (each comprising an instance type and Availability Zone) in an EC2 Fleet or Spot Fleet, you can use an *allocation strategy* to manage how Amazon EC2 fulfills your Spot and On-Demand capacities from these pools. The allocation strategies can optimize for available capacity, price, and the instance types to use. There are different allocation strategies for Spot Instances and On-Demand Instances. **Topics** + [ ## Allocation strategies for Spot Instances ](#ec2-fleet-allocation-strategies-for-spot-instances) + [ ## Allocation strategies for On-Demand Instances ](#ec2-fleet-allocation-strategies-for-on-demand-instances) + [ ## Choose the appropriate Spot allocation strategy ](#ec2-fleet-allocation-use-cases) + [ ## Maintain target capacity for Spot Instances ](#ec2-fleet-maintain-fleet-capacity) + [ ## Prioritize instance types for On-Demand capacity ](#ec2-fleet-on-demand-priority) ## Allocation strategies for Spot Instances Your launch configuration determines all the possible Spot capacity pools (instance types and Availability Zones) from which EC2 Fleet or Spot Fleet can launch Spot Instances. However, when launching instances, the fleet uses the allocation strategy that you specify to pick the specific pools from all your possible pools. **Note** (Linux instances only) If you configure your Spot Instance to launch with [AMD SEV-SNP](sev-snp.md) turned on, you are charged an additional hourly usage fee that is equivalent to 10% of the [On-Demand hourly rate](https://aws.amazon.com/ec2/pricing/on-demand/) for the selected instance type. If the allocation strategy uses price as an input, the fleet does not include this additional fee; only the Spot price is used. You can specify one of the following allocation strategies for Spot Instances: **Price capacity optimized** (recommended) The fleet identifies the pools with the highest capacity availability for the number of instances that are launching. This means that we will request Spot Instances from the pools that we believe have the lowest chance of interruption in the near term. The fleet then requests Spot Instances from the lowest priced of these pools. The **price capacity optimized** allocation strategy is the best choice for most Spot workloads, such as stateless containerized applications, microservices, web applications, data and analytics jobs, and batch processing. If you're using the AWS CLI, the parameter name is `price-capacity-optimized` for EC2 Fleet and `priceCapacityOptimized` for Spot Fleet. **Capacity optimized** The fleet identifies the pools with the highest capacity availability for the number of instances that are launching. This means that we will request Spot Instances from the pools that we believe have the lowest chance of interruption in the near term. With Spot Instances, pricing changes slowly over time based on long-term trends in supply and demand, but capacity fluctuates in real time. The **capacity optimized** strategy automatically launches Spot Instances into the most available pools by looking at real-time capacity data and predicting which are the most available. This works well for workloads that may have a higher cost of interruption associated with restarting work, such as long Continuous Integration (CI), image and media rendering, Deep Learning, and High Performance Compute (HPC) workloads that may have a higher cost of interruption associated with restarting work. By offering the possibility of fewer interruptions, the **capacity optimized** strategy can lower the overall cost of your workload. Alternatively, you can use the **capacity optimized prioritized** allocation strategy with a priority parameter to order instance types from highest to lowest priority. You can set the same priority for different instance types. The fleet will optimize for capacity first, but will honor instance type priorities on a best-effort basis (for example, if honoring the priorities will not significantly affect the fleet's ability to provision optimal capacity). This is a good option for workloads where the possibility of disruption must be minimized and the preference for certain instance types matters. Note that when you set the priority for instance types for your Spot capacity, the same priority is also applied to your On-Demand Instances if the On-Demand allocation strategy is set to **prioritized**. For Spot Fleet, using priorities is supported only if your fleet uses a launch template. If you're using the AWS CLI, the parameter names are `capacity-optimized` and `capacity-optimized-prioritized` for EC2 Fleet and `capacityOptimized` and `capacityOptimizedPrioritized` for Spot Fleet. **Diversified** The Spot Instances are distributed across all Spot capacity pools. If you're using the AWS CLI, the parameter name is `diversified` for both EC2 Fleet and Spot Fleet. **Lowest price** (not recommended) We don't recommend the **lowest price** allocation strategy because it has the highest risk of interruption for your Spot Instances. The Spot Instances come from the lowest priced pool that has available capacity. When using the AWS CLI, this is the default strategy. However, we recommend that you override the default by specifying the **price capacity optimized** allocation strategy. With the lowest price strategy, if the lowest priced pool doesn't have available capacity, the Spot Instances come from the next lowest priced pool that has available capacity. If a pool runs out of capacity before fulfilling your desired capacity, the fleet will continue to fulfill your request by drawing from the next lowest priced pool. To ensure that your desired capacity is met, you might receive Spot Instances from several pools. Because this strategy only considers instance price and not capacity availability, it might lead to high interruption rates. The lowest price allocation strategy is only available when using the AWS CLI. The parameter name is `lowest-price` for EC2 Fleet and `lowestPrice` for Spot Fleet. **Number of pools to use** The number of Spot pools across which to allocate your target Spot capacity. Valid only when the allocation strategy is set to **lowest price**. The fleet selects the lowest priced Spot pools and evenly allocates your target Spot capacity across the number of Spot pools that you specify. Note that the fleet attempts to draw Spot Instances from the number of pools that you specify on a best effort basis. If a pool runs out of Spot capacity before fulfilling your target capacity, the fleet will continue to fulfill your request by drawing from the next lowest priced pool. To ensure that your target capacity is met, you might receive Spot Instances from more than the number of pools that you specified. Similarly, if most of the pools have no Spot capacity, you might receive your full target capacity from fewer than the number of pools that you specified. This parameter is only available when specifying the **lowest price** allocation strategy and only when using the AWS CLI. The parameter name is `InstancePoolsToUseCount` for both EC2 Fleet and Spot Fleet. ## Allocation strategies for On-Demand Instances Your launch configuration determines all the possible capacity pools (instance types and Availability Zones) from which EC2 Fleet or Spot Fleet can launch On-Demand Instances. However, when launching instances, the fleet uses the allocation strategy that you specify to pick the specific pools from all your possible pools. You can specify one of the following allocation strategies for On-Demand Instances: **Lowest price** The On-Demand Instances come from the lowest priced pool that has available capacity. This is the default strategy. If the lowest priced pool doesn't have available capacity, the On-Demand Instances come from the next lowest priced pool that has available capacity. If a pool runs out of capacity before fulfilling your desired capacity, the fleet will continue to fulfill your request by drawing from the next lowest priced pool. To ensure that your desired capacity is met, you might receive On-Demand Instances from several pools. **Prioritized** The fleet uses the priority that you assigned to each launch template override, launching instance types in order of the highest priority first. This strategy can't be used with attribute-based instance type selection. For an example of how to use this allocation strategy, see [Prioritize instance types for On-Demand capacity](#ec2-fleet-on-demand-priority). ## Choose the appropriate Spot allocation strategy You can optimize your fleet for your use case by choosing the appropriate Spot allocation strategy. ### Balance lowest price and capacity availability To balance the trade-offs between the lowest priced Spot capacity pools and the Spot capacity pools with the highest capacity availability, we recommend that you use the **price capacity optimized** allocation strategy. This strategy makes decisions about which pools to request Spot Instances from based on both the price of the pools and the capacity availability of Spot Instances in those pools. This means that we will request Spot Instances from the pools that we believe have the lowest chance of interruption in the near term, while still taking price into consideration. If your fleet runs resilient and stateless workloads, including containerized applications, microservices, web applications, data and analytics jobs, and batch processing, then use the **price capacity optimized** allocation strategy for optimal cost savings and capacity availability. If your fleet runs workloads that might have a higher cost of interruption associated with restarting work, then you should implement checkpointing so that applications can restart from that point if they're interrupted. By using checkpointing, you make the **price capacity optimized** allocation strategy a good fit for these workloads because it allocates capacity from the lowest priced pools that also offer a low Spot Instance interruption rate. For example JSON configurations that use the **price capacity optimized** allocation strategy, see the following: + EC2 Fleet – [Example 10: Launch Spot Instances in a price-capacity-optimized fleet](ec2-fleet-examples.md#ec2-fleet-config11) + Spot Fleet – [Example 11: Launch Spot Instances in a priceCapacityOptimized fleet](spot-fleet-examples.md#fleet-config11) ### When workloads have a high cost of interruption You can optionally use the **capacity optimized** strategy if you run workloads that either use similarly priced instance types, or where the cost of interruption is so significant that any cost saving is inadequate in comparison to a marginal increase in interruptions. This strategy allocates capacity from the most available Spot capacity pools that offer the possibility of fewer interruptions, which can lower the overall cost of your workload. When the possibility of interruptions must be minimized but the preference for certain instance types matters, you can express your pool priorities by using the **capacity optimized prioritized** allocation strategy and then setting the order of instance types to use from highest to lowest priority. Note that when you set priorities for **capacity optimized prioritized**, the same priorities are also applied to your On-Demand Instances if the On-Demand allocation strategy is set to **prioritized**. Also note that, for Spot Fleet, using priorities is supported only if your fleet uses a launch template. For example JSON configurations that use the **capacity optimized** allocation strategy, see the following: + EC2 Fleet – [Example 8: Launch Spot Instances in a capacity-optimized fleet](ec2-fleet-examples.md#ec2-fleet-config9) + Spot Fleet – [Example 9: Launch Spot Instances in a capacity-optimized fleet](spot-fleet-examples.md#fleet-config9) For example JSON configurations that use the **capacity optimized prioritized** allocation strategy, see the following: + EC2 Fleet – [Example 9: Launch Spot Instances in a capacity-optimized fleet with priorities](ec2-fleet-examples.md#ec2-fleet-config10) + Spot Fleet – [Example 10: Launch Spot Instances in a capacity-optimized fleet with priorities](spot-fleet-examples.md#fleet-config10) ### When your workload is time flexible and capacity availability is not a factor If your fleet is small or runs for a short time, you can use **price capacity optimized** to maximize cost savings while still considering capacity availability. ### When your fleet is large or runs for a long time If your fleet is large or runs for a long time, you can improve the availability of your fleet by distributing the Spot Instances across multiple pools using the **diversified** strategy. For example, if your fleet specifies 10 pools and a target capacity of 100 instances, the fleet launches 10 Spot Instances in each pool. If the Spot price for one pool exceeds your maximum price for this pool, only 10% of your fleet is affected. Using this strategy also makes your fleet less sensitive to increases in the Spot price in any one pool over time. With the **diversified** strategy, the fleet does not launch Spot Instances into any pools with a Spot price that is equal to or higher than the [On-Demand price](https://aws.amazon.com/ec2/pricing/). ## Maintain target capacity for Spot Instances After Spot Instances are terminated due to a change in the Spot price or available capacity of a Spot capacity pool, a fleet of type `maintain` launches replacement Spot Instances. The allocation strategy determines the pools from which the replacement instances are launched, as follows: + If the allocation strategy is **price capacity optimized**, the fleet launches replacement instances in the pools that have the most Spot Instance capacity availability while also taking price into consideration and identifying lowest priced pools with high capacity availability. + If the allocation strategy is **capacity optimized**, the fleet launches replacement instances in the pools that have the most Spot Instance capacity availability. + If the allocation strategy is **diversified**, the fleet distributes the replacement Spot Instances across the remaining pools. ## Prioritize instance types for On-Demand capacity When an EC2 Fleet or Spot Fleet attempts to fulfill your On-Demand capacity, it defaults to launching the lowest priced instance type first. If the On-Demand allocation strategy is set to **prioritized**, the fleet uses priority to determine which instance type to use first when fulfilling On-Demand capacity. The priority is assigned to the launch template override, and the highest priority is launched first. **Example: Prioritize instance types** In this example, you configure three launch template overrides, each with a different instance type. The On-Demand price for the instance types range in price. The following are the instance types used in this example, listed in order of price, starting with the least expensive instance type: + `m4.large` – least expensive + `m5.large` + `m5a.large` If you do not use priority to determine the order, the fleet fulfills the On-Demand capacity by starting with the least expensive instance type. However, say you have unused `m5.large` Reserved Instances that you want to use first. You can set the launch template override priority so that the instance types are used in the order of priority, as follows: + `m5.large` – priority 1 + `m4.large` – priority 2 + `m5a.large` – priority 3 # Use Capacity Rebalancing in EC2 Fleet and Spot Fleet to replace at-risk Spot Instances With Capacity Rebalancing, your EC2 Fleet or Spot Fleet can maintain the desired Spot capacity by proactively replacing Spot Instances at risk of interruption. When a Spot Instance is at an elevated risk of interruption, Amazon EC2 sends a [rebalance recommendation](rebalance-recommendations.md). If Capacity Rebalancing is enabled, the rebalance recommendation triggers the launch of a new Spot Instance before the at-risk instance is interrupted. Capacity Rebalancing helps you maintain workload availability by proactively augmenting your fleet with new Spot Instances before running instances are interrupted by Amazon EC2. **To configure EC2 Fleet to use Capacity Rebalancing to launch a replacement Spot Instance** Use the [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command and the relevant parameters in the `MaintenanceStrategies` structure. For an example JSON configuration, see [Example 7: Configure Capacity Rebalancing to launch replacement Spot Instances](ec2-fleet-examples.md#ec2-fleet-config8). **To configure Spot Fleet to use Capacity Rebalancing to launch a replacement Spot Instance** You can use the Amazon EC2 console or the AWS CLI to configure Capacity Rebalancing. (Console) When creating the Spot Fleet, select the **Capacity rebalance** checkbox. For more information, see step 6.d. in [Create a Spot Fleet request using defined parameters](create-spot-fleet.md#create-spot-fleet-advanced). (AWS CLI) Use the [request-spot-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/request-spot-fleet.html) command and the relevant parameters in the `SpotMaintenanceStrategies` structure. For an example JSON configuration, see [Example 8: Configure Capacity Rebalancing to launch replacement Spot Instances](spot-fleet-examples.md#fleet-config8). **Topics** + [ ## Limitations ](#ec2-fleet-capacity-rebalance-limitations) + [ ## Configuration options ](#ec2-fleet-capacity-rebalance-config-options) + [ ## Considerations ](#ec2-fleet-capacity-rebalance-considerations) ## Limitations + Capacity Rebalancing is available only for fleets of type `maintain`. + When the fleet is running, you can't modify the Capacity Rebalancing setting. To change the Capacity Rebalancing setting, you must delete the fleet and create a new fleet. ## Configuration options The `ReplacementStrategy` for EC2 Fleet and Spot Fleet supports the following two values: `launch-before-terminate` Amazon EC2 terminates the Spot Instances that receive a rebalance notification after new replacement Spot Instances are launched. When you specify `launch-before-terminate`, you must also specify a value for `termination-delay`. After the new replacement instances are launched, Amazon EC2 waits for the duration of the `termination-delay`, and then terminates the old instances. For `termination-delay`, the minimum is 120 seconds (2 minutes), and the maximum is 7200 seconds (2 hours). We recommend that you use `launch-before-terminate` only if you can predict how long your instance shutdown procedures will take to complete. This will ensure that the old instances are terminated only after the shutdown procedures are completed. Note that Amazon EC2 can interrupt the old instances with a two-minute warning before the `termination-delay`. We strongly recommend against using the `lowest-price` (EC2 Fleet) or `lowestPrice` (Spot Fleet) allocation strategy in combination with `launch-before-terminate` to avoid having replacement Spot Instances that are also at an elevated risk of interruption. `launch` Amazon EC2 launches replacement Spot Instances when a rebalance notification is emitted for existing Spot Instances. Amazon EC2 does not terminate the instances that receive a rebalance notification. You can terminate the old instances, or you can leave them running. You are charged for all instances while they are running. ## Considerations If you configure an EC2 Fleet or Spot Fleet for Capacity Rebalancing, consider the following: **Provide as many Spot capacity pools in the request as possible** Configure your fleet to use multiple instance types and Availability Zones. This provides the flexibility to launch Spot Instances in various Spot capacity pools. For more information, see [Be flexible about instance types and Availability Zones](spot-best-practices.md#be-instance-type-flexible). **Avoid an elevated risk of interruption of replacement Spot Instances** To avoid an elevated risk of interruption, we recommend the `capacity-optimized` or `capacity-optimized-prioritized` allocation strategy. These strategies ensure that replacement Spot Instances are launched in the most optimal Spot capacity pools, and are therefore less likely to be interrupted in the near future. For more information, see [Use the price and capacity optimized allocation strategy](spot-best-practices.md#use-capacity-optimized-allocation-strategy). If you use the `lowest-price` allocation strategy, your replacement Spot Instances might be at an elevated risk of interruption. This is because Amazon EC2 will always launch instances in the lowest priced pool that has available capacity at that moment, even if your replacement Spot Instances are likely to be interrupted soon after being launched. **Amazon EC2 will only launch a new instance if availability is the same or better** One of the goals of Capacity Rebalancing is to improve a Spot Instance's availability. If an existing Spot Instance receives a rebalance recommendation, Amazon EC2 will only launch a new instance if the new instance provides the same or better availability than the existing instance. If the risk of interruption of a new instance will be worse than the existing instance, then Amazon EC2 will not launch a new instance. Amazon EC2 will, however, continue to assess the Spot capacity pools, and will launch a new instance if availability improves. There is a chance that your existing instance will be interrupted without Amazon EC2 proactively launching a new instance. When this happens, Amazon EC2 will attempt to launch a new instance regardless of whether the new instance has a high risk of interruption. **Capacity Rebalancing does not increase your Spot Instance interruption rate** When you enable Capacity Rebalancing, it does not increase your [Spot Instance interruption rate](spot-interruptions.md) (the number of Spot Instances that are reclaimed when Amazon EC2 needs the capacity back). However, if Capacity Rebalancing detects an instance is at risk of interruption, Amazon EC2 will immediately attempt to launch a new instance. The result is that more instances might be replaced than if you’d waited for Amazon EC2 to launch a new instance after the at-risk instance was interrupted. While you might replace more instances with Capacity Rebalancing enabled, you benefit from being proactive rather than reactive by having more time to take action before your instances are interrupted. With a [Spot Instance interruption notice](spot-instance-termination-notices.md), you typically only have up to two minutes to gracefully shut down your instance. With Capacity Rebalancing launching a new instance in advance, you give existing processes a better chance of completing on your at-risk instance, you can start your instance shutdown procedures, and prevent new work from being scheduled on your at-risk instance. You can also start preparing the newly-launched instance to take over the application. With Capacity Rebalancing’s proactive replacement, you benefit from graceful continuity. As a theoretical example to demonstrate the risks and benefits of using Capacity Rebalancing, consider the following scenario: + 2:00 PM – A rebalance recommendation is received for instance-A, and Amazon EC2 immediately starts attempting to launch a replacement instance-B, giving you time to start your shutdown procedures.\$1 + 2:30 PM – A rebalance recommendation is received for instance-B, replaced with instance-C, giving you time to start your shutdown procedures.\$1 + 2:32 PM – If Capacity Rebalancing wasn’t enabled, and if a Spot Instance interruption notice would've been received at 2:32 PM for instance-A, you would only have had up to two minutes to take action, but Instance-A would have been running up till this time. \$1 If `launch-before-terminate` is specified, Amazon EC2 will terminate the at-risk instance after the replacement instance comes online. **Amazon EC2 can launch new replacement Spot Instances until fulfilled capacity is double target capacity** When a fleet is configured for Capacity Rebalancing, the fleet attempts to launch a new replacement Spot Instance for every Spot Instance that receives a rebalance recommendation. After a Spot Instance receives a rebalance recommendation, it is no longer counted as part of the fulfilled capacity. Depending on the replacement strategy, Amazon EC2 either terminates the instance after a preconfigured termination delay, or leaves it running. This gives you the opportunity to perform [rebalancing actions](rebalance-recommendations.md#rebalancing-actions) on the instance. If your fleet reaches double its target capacity, it stops launching new replacement instances even if the replacement instances themselves receive a rebalance recommendation. For example, you create a fleet with a target capacity of 100 Spot Instances. All of the Spot Instances receive a rebalance recommendation, which causes Amazon EC2 to launch 100 replacement Spot Instances. This raises the number of fulfilled Spot Instances to 200, which is double the target capacity. Some of the replacement instances receive a rebalance recommendation, but no more replacement instances are launched because the fleet can't exceed double its target capacity. Note that you are charged for all of the instances while they are running. **We recommend that you configure your fleet to terminate Spot Instances that receive a rebalance recommendation** If you configure your fleet for Capacity Rebalancing, we recommend that you choose `launch-before-terminate` with an appropriate termination delay only if you can predict how long your instance shutdown procedures will take to complete. This will ensure that the old instances are terminated only after the shutdown procedures are completed. If you choose to terminate the instances that are recommended for rebalance yourself, we recommend that you monitor the rebalance recommendation signal that is received by the Spot Instances in the fleet. By monitoring the signal, you can quickly perform [rebalancing actions](rebalance-recommendations.md#rebalancing-actions) on the affected instances before Amazon EC2 interrupts them, and then you can manually terminate them. If you do not terminate the instances, you continue paying for them while they are running. Amazon EC2 does not automatically terminate the instances that receive a rebalance recommendation. You can set up notifications using Amazon EventBridge or instance metadata. For more information, see [Monitor rebalance recommendation signals](rebalance-recommendations.md#monitor-rebalance-recommendations). **The fleet does not count instances that receive a rebalance recommendation when calculating fulfilled capacity during scale in or out** If your fleet is configured for Capacity Rebalancing, and you change the target capacity to either scale in or scale out, the fleet does not count the instances that are marked for rebalance as part of the fulfilled capacity, as follows: + Scale in – If you decrease your desired target capacity, Amazon EC2 terminates instances that are not marked for rebalance until the desired capacity is reached. The instances that are marked for rebalance are not counted towards the fulfilled capacity. For example, you create a fleet with a target capacity of 100 Spot Instances. 10 instances receive a rebalance recommendation, so Amazon EC2 launches 10 new replacement instances, resulting in a fulfilled capacity of 110 instances. You then reduce the target capacity to 50 (scale in), but the fulfilled capacity is actually 60 instances because the 10 instances that are marked for rebalance are not terminated by Amazon EC2. You need to manually terminate these instances, or you can leave them running. + Scale out – If you increase your desired target capacity, Amazon EC2 launches new instances until the desired capacity is reached. The instances that are marked for rebalance are not counted towards the fulfilled capacity. For example, you create a fleet with a target capacity of 100 Spot Instances. 10 instances receive a rebalance recommendation, so the fleet launches 10 new replacement instances, resulting in a fulfilled capacity of 110 instances. You then increase the target capacity to 200 (scale out), but the fulfilled capacity is actually 210 instances because the 10 instances that are marked for rebalance are not counted by the fleet as part of the target capacity. You need to manually terminate these instances, or you can leave them running. # Use Capacity Reservations to reserve On-Demand capacity in EC2 Fleet With On-Demand Capacity Reservations, you can reserve compute capacity for your On-Demand Instances in a specified Availability Zone for any duration. You can configure an EC2 Fleet to use the Capacity Reservations first when launching On-Demand Instances. On-Demand Capacity Reservations are available only for EC2 Fleet with the request type set to `instant`. Capacity Reservations are configured as either `open` or `targeted`. EC2 Fleet can launch On-Demand Instances into either `open` or `targeted` Capacity Reservations, as follows: + If a Capacity Reservation is `open`, On-Demand Instances that have matching attributes automatically run in the reserved capacity. + If a Capacity Reservation is `targeted`, On-Demand Instances must specifically target it to run in the reserved capacity. This is useful for using up specific Capacity Reservations or for controlling when to use specific Capacity Reservations. If you use `targeted` Capacity Reservations in your EC2 Fleet, there must be enough Capacity Reservations to fulfil the target On-Demand capacity, otherwise the launch fails. To avoid a launch fail, rather add the `targeted` Capacity Reservations to a resource group, and then target the resource group. The resource group doesn't need to have enough Capacity Reservations; if it runs out of Capacity Reservations before the target On-Demand capacity is fulfilled, the fleet can launch the remaining target capacity into regular On-Demand capacity. **To use Capacity Reservations with EC2 Fleet** 1. Configure the fleet as type `instant`. You can't use Capacity Reservations for fleets of other types. 1. Configure the usage strategy for Capacity Reservations as `use-capacity-reservations-first`. 1. In the launch template, for **Capacity reservation**, choose either **Open** or **Target by group**. If you choose **Target by group**, specify the Capacity Reservations resource group ID. When the fleet attempts to fulfil the On-Demand capacity, if it finds that multiple instance pools have unused matching Capacity Reservations, it determines the pools in which to launch the On-Demand Instances based on the On-Demand allocation strategy (`lowest-price` or `prioritized`). **Related resources** + For CLI examples of how to configure a fleet to use Capacity Reservations to fulfil On-Demand capacity, see [Example CLI configurations for EC2 Fleet](ec2-fleet-examples.md), specifically Examples 5 through 7. + For a tutorial that takes you through the steps for creating Capacity Reservations, using them in your fleet, and viewing how many Capacity Reservations are remaining, see [Tutorial: Configure EC2 Fleet to launch On-Demand Instances using targeted Capacity Reservations](ec2-fleet-launch-on-demand-instances-using-targeted-capacity-reservations-walkthrough.md) + For information about configuring Capacity Reservations, see [Reserve compute capacity with EC2 On-Demand Capacity Reservations](ec2-capacity-reservations.md) and the [On-Demand Capacity Reservation FAQs](https://aws.amazon.com/ec2/faqs/#On-Demand_Capacity_Reservation). # Work with EC2 Fleet To start using an EC2 Fleet, create a request that includes the total target capacity, On-Demand capacity, Spot capacity, and a launch template specifying the configuration for the instances in the fleet. You can optionally specify additional parameters, or let the fleet use default values. You can also tag the fleet request, and its instances and volumes, when you create the fleet. The fleet launches On-Demand Instances when there is available capacity, and launches Spot Instances when your maximum price exceeds the Spot price and capacity is available. Once your fleet is launched, you can describe the fleet request, the instances in the fleet, and any fleet events. You can also assign additional tags as needed. If you need to change any fleet parameters, such as the total target capacity, you can modify the fleet, provided it was configured to maintain capacity. You can't modify the capacity of a one-time request after it's been submitted. The fleet request remains active until it expires or you delete it. When you delete the fleet request, you can either terminate the instances or leave them running. If you choose to leave them running, the On-Demand Instances run until you terminate them, and the Spot Instances run until they're interrupted or you terminate them. **Topics** + [ # EC2 Fleet request states ](EC2-fleet-states.md) + [ # EC2 Fleet prerequisites ](ec2-fleet-prerequisites.md) + [ # Create an EC2 Fleet ](create-ec2-fleet.md) + [ # Tag a new or existing EC2 Fleet request and the instances and volumes it launches ](tag-ec2-fleet.md) + [ # Describe an EC2 Fleet, its instances, and its events ](describe-ec2-fleet.md) + [ # Modify an EC2 Fleet ](modify-ec2-fleet.md) + [ # Delete an EC2 Fleet request and the instances in the fleet ](delete-fleet.md) # EC2 Fleet request states An EC2 Fleet request can be one of various states, with each state indicating a different stage of the request's lifecycle and its management of instances. An EC2 Fleet request can be in one of the following states: `submitted` The EC2 Fleet request is being evaluated and Amazon EC2 is preparing to launch the target number of instances. If a request would exceed your fleet limits, it is deleted immediately. `active` The EC2 Fleet request has been validated and Amazon EC2 is attempting to maintain the target number of running instances. The request remains in this state until it is modified or deleted. `modifying` The EC2 Fleet request is being modified. The request remains in this state until the modification is fully processed or the request is deleted. Only a `maintain` fleet type can be modified. This state does not apply to other request types. `deleted_running` The EC2 Fleet request is deleted and does not launch additional Spot Instances. Its existing instances continue to run until they are interrupted or terminated manually. The request remains in this state until all instances are interrupted or terminated. Only an EC2 Fleet of type `maintain` or `request` can have running instances after the EC2 Fleet request is deleted. A deleted `instant` fleet with running instances is not supported. This state does not apply to `instant` fleets. `deleted_terminating` The EC2 Fleet request is deleted and its instances are terminating. The request remains in this state until all instances are terminated. `deleted` The EC2 Fleet request is deleted and has no running instances. The request is deleted two days after its instances are terminated. # EC2 Fleet prerequisites **Topics** + [ ## Launch template ](#ec2-fleet-prerequisites-launch-template) + [ ## Service-linked role for EC2 Fleet ](#ec2-fleet-service-linked-role) + [ ## Grant access to customer managed keys for use with encrypted AMIs and EBS snapshots ](#ec2-fleet-service-linked-roles-access-to-cmks) + [ ## Permissions for EC2 Fleet users ](#ec2-fleet-iam-users) ## Launch template A launch template specifies the configuration information about the instances to launch, such as the instance type and Availability Zone. For more information about launch templates, see [Store instance launch parameters in Amazon EC2 launch templates](ec2-launch-templates.md). ## Service-linked role for EC2 Fleet The `AWSServiceRoleForEC2Fleet` role grants the EC2 Fleet permission to request, launch, terminate, and tag instances on your behalf. Amazon EC2 uses this service-linked role to complete the following actions: + `ec2:RunInstances` – Launch instances. + `ec2:RequestSpotInstances` – Request Spot Instances. + `ec2:TerminateInstances` – Terminate instances. + `ec2:DescribeImages` – Describe Amazon Machine Images (AMIs) for the instances. + `ec2:DescribeInstanceStatus` – Describe the status of the instances. + `ec2:DescribeSubnets` – Describe the subnets for instances. + `ec2:CreateTags` – Add tags to the EC2 Fleet, instances, and volumes. Ensure that this role exists before you use the AWS CLI or an API to create an EC2 Fleet. **Note** An `instant` EC2 Fleet does not require this role. To create the role, use the IAM console as follows. **To create the AWSServiceRoleForEC2Fleet role for EC2 Fleet** 1. Open the IAM console at [https://console.aws.amazon.com/iam/](https://console.aws.amazon.com/iam/). 1. In the navigation pane, choose **Roles**. 1. Choose **Create role**. 1. On the **Select trusted entity** page, do the following: 1. For **Trusted entity type**, choose **AWS service**. 1. Under **Use case**, for **Service or use case**, choose **EC2 - Fleet**. **Tip** Be sure to choose **EC2 - Fleet**. If you choose **EC2**, the **EC2 - Fleet** use case does not appear in the **Use case** list. The **EC2 - Fleet** use case will automatically create a policy with the required IAM permissions and will suggest **AWSServiceRoleForEC2Fleet** as the role name. 1. Choose **Next**. 1. On the **Add permissions page**, choose **Next**. 1. On the **Name, review, and create** page, choose **Create role**. If you no longer need to use EC2 Fleet, we recommend that you delete the **AWSServiceRoleForEC2Fleet** role. After this role is deleted from your account, you can create the role again if you create another fleet. For more information, see [Service-linked roles](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_create-service-linked-role.html) in the *IAM User Guide*. ## Grant access to customer managed keys for use with encrypted AMIs and EBS snapshots If you specify an [encrypted AMI](AMIEncryption.md) or an encrypted Amazon EBS snapshot in your EC2 Fleet and you use an AWS KMS key for encryption, you must grant the **AWSServiceRoleForEC2Fleet** role permission to use the customer managed key so that Amazon EC2 can launch instances on your behalf. To do this, you must add a grant to the customer managed key, as shown in the following procedure. When providing permissions, grants are an alternative to key policies. For more information, see [Using grants](https://docs.aws.amazon.com/kms/latest/developerguide/grants.html) and [Using key policies in AWS KMS](https://docs.aws.amazon.com/kms/latest/developerguide/key-policies.html) in the *AWS Key Management Service Developer Guide*. **To grant the AWSServiceRoleForEC2Fleet role permissions to use the customer managed key** + Use the [create-grant](https://docs.aws.amazon.com/cli/latest/reference/kms/create-grant.html) command to add a grant to the customer managed key and to specify the principal (the **AWSServiceRoleForEC2Fleet** service-linked role) that is given permission to perform the operations that the grant permits. The customer managed key is specified by the `key-id` parameter and the ARN of the customer managed key. The principal is specified by the `grantee-principal` parameter and the ARN of the **AWSServiceRoleForEC2Fleet** service-linked role. ``` aws kms create-grant \ --region us-east-1 \ --key-id arn:aws:kms:us-east-1:444455556666:key/1234abcd-12ab-34cd-56ef-1234567890ab \ --grantee-principal arn:aws:iam::111122223333:role/AWSServiceRoleForEC2Fleet \ --operations "Decrypt" "Encrypt" "GenerateDataKey" "GenerateDataKeyWithoutPlaintext" "CreateGrant" "DescribeKey" "ReEncryptFrom" "ReEncryptTo" ``` ## Permissions for EC2 Fleet users If your users will create or manage an EC2 Fleet, be sure to grant them the required permissions. **To create a policy for EC2 Fleet** 1. Open the IAM console at [https://console.aws.amazon.com/iam/](https://console.aws.amazon.com/iam/). 1. In the navigation pane, choose **Policies**. 1. Choose **Create policy**. 1. On the **Create policy** page, choose the **JSON** tab, replace the text with the following, and choose **Review policy**. ------ #### [ JSON ] **** ``` { "Version":"2012-10-17", "Statement": [ { "Effect": "Allow", "Action": [ "ec2:*" ], "Resource": "*" }, { "Effect": "Allow", "Action": [ "iam:ListRoles", "iam:PassRole", "iam:ListInstanceProfiles" ], "Resource":"arn:aws:iam::123456789012:role/DevTeam*" } ] } ``` ------ The `ec2:*` grants a user permission to call all Amazon EC2 API actions. To limit the user to specific Amazon EC2 API actions, specify those actions instead. The user must have permission to call the `iam:ListRoles` action to enumerate existing IAM roles, the `iam:PassRole` action to specify the EC2 Fleet role, and the `iam:ListInstanceProfiles` action to enumerate existing instance profiles. (Optional) To enable a user to create roles or instance profiles using the IAM console, you must also add the following actions to the policy: + `iam:AddRoleToInstanceProfile` + `iam:AttachRolePolicy` + `iam:CreateInstanceProfile` + `iam:CreateRole` + `iam:GetRole` + `iam:ListPolicies` 1. On the **Review policy** page, enter a policy name and description, and choose **Create policy**. 1. To provide access, add permissions to your users, groups, or roles: + Users and groups in AWS IAM Identity Center: Create a permission set. Follow the instructions in [Create a permission set](https://docs.aws.amazon.com//singlesignon/latest/userguide/howtocreatepermissionset.html) in the *AWS IAM Identity Center User Guide*. + Users managed in IAM through an identity provider: Create a role for identity federation. Follow the instructions in [Create a role for a third-party identity provider (federation)](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_roles_create_for-idp.html) in the *IAM User Guide*. + IAM users: + Create a role that your user can assume. Follow the instructions in [Create a role for an IAM user](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_roles_create_for-user.html) in the *IAM User Guide*. + (Not recommended) Attach a policy directly to a user or add a user to a user group. Follow the instructions in [Adding permissions to a user (console)](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_users_change-permissions.html#users_change_permissions-add-console) in the *IAM User Guide*. # Create an EC2 Fleet To create an EC2 Fleet, define the fleet configuration in a JSON file and reference the file with the [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command. In the JSON file, you must specify the total target capacity for the fleet, separate target capacities for Spot Instances and On-Demand Instances, and a launch template that defines the configuration for the instances in the fleet, such as an AMI, instance type, subnet or Availability Zone, and one or more security groups. You can optionally specify additional configurations, such as parameters to override the launch template configuration, allocation strategies for selecting Spot Instances and On-Demand Instances from the EC2 capacity pools, and the maximum amount you're willing to pay for the fleet. For more information, see [Configuration options for your EC2 Fleet or Spot Fleet](ec2-fleet-configuration-strategies.md). The EC2 Fleet launches On-Demand Instances when capacity is available, and launches Spot Instances when your maximum price exceeds the Spot price and capacity is available. If your fleet includes Spot Instances and is of type `maintain`, Amazon EC2 will attempt to maintain your fleet target capacity when your Spot Instances are interrupted. ## EC2 Fleet limitations The following limitations apply to EC2 Fleet: + Creating an EC2 Fleet is available only through the [Amazon EC2 API](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_CreateFleet.html), [AWS CLI](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html), [AWS SDKs](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_CreateFleet.html#API_CreateFleet_SeeAlso), and [CloudFormation](https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-ec2-ec2fleet.html). + An EC2 Fleet request can't span AWS Regions. You need to create a separate EC2 Fleet for each Region. + An EC2 Fleet request can't span different subnets from the same Availability Zone. ## Create an EC2 Fleet To launch a fleet of instances using EC2 Fleet, you need only specify the following parameters in your fleet request, and the fleet will use the default values for the other parameters: + `LaunchTemplateId` or `LaunchTemplateName` – Specifies the launch template to use (which contains the parameters for the instances to launch, such as the instance type and Availability Zone) + `TotalTargetCapacity` – Specifies the total target capacity for the fleet + `DefaultTargetCapacityType` – Specifies whether the default purchasing option is On-Demand or Spot To override the parameters specified in the launch template, you can specify one or more overrides. Each override can vary by instance type, Availability Zone, subnet, and maximum price, and can include a different weighted capacity. As an alternative to specifying an instance type, you can specify the attributes that an instance must have, and Amazon EC2 will identify all the instance types with those attributes. For more information, see [Specify attributes for instance type selection for EC2 Fleet or Spot Fleet](ec2-fleet-attribute-based-instance-type-selection.md). For EC2 Fleets of type `instant`, you can specify a Systems Manager parameter instead of the AMI ID. You can specify the Systems Manager parameter in the override or in the launch template. For more information, see [Use a Systems Manager parameter instead of an AMI ID](create-launch-template.md#use-an-ssm-parameter-instead-of-an-ami-id). You can specify the fleet parameters in a JSON file. For information about all the possible parameters you can specify, see [View all the EC2 Fleet configuration options](#ec2-fleet-cli-skeleton). For fleet configuration examples, see [Example CLI configurations for EC2 Fleet](ec2-fleet-examples.md). There is currently no console support for creating an EC2 Fleet. **To create an EC2 Fleet** Use the [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command to create the fleet and specify the JSON file that contains the fleet configuration parameters. ``` aws ec2 create-fleet --cli-input-json file://file_name.json ``` The following is example output for a fleet of type `request` or `maintain`. ``` { "FleetId": "fleet-12a34b55-67cd-8ef9-ba9b-9208dEXAMPLE" } ``` The following is example output for a fleet of type `instant` that launched the target capacity. ``` { "FleetId": "fleet-12a34b55-67cd-8ef9-ba9b-9208dEXAMPLE", "Errors": [], "Instances": [ { "LaunchTemplateAndOverrides": { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-01234a567b8910abcEXAMPLE", "Version": "1" }, "Overrides": { "InstanceType": "c5.large", "AvailabilityZone": "us-east-1a" } }, "Lifecycle": "on-demand", "InstanceIds": [ "i-1234567890abcdef0", "i-9876543210abcdef9" ], "InstanceType": "c5.large", "Platform": null }, { "LaunchTemplateAndOverrides": { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-01234a567b8910abcEXAMPLE", "Version": "1" }, "Overrides": { "InstanceType": "c4.large", "AvailabilityZone": "us-east-1a" } }, "Lifecycle": "on-demand", "InstanceIds": [ "i-5678901234abcdef0", "i-5432109876abcdef9" ] ] } ``` The following is example output for a fleet of type `instant` that launched part of the target capacity with errors for instances that were not launched. ``` { "FleetId": "fleet-12a34b55-67cd-8ef9-ba9b-9208dEXAMPLE", "Errors": [ { "LaunchTemplateAndOverrides": { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-01234a567b8910abcEXAMPLE", "Version": "1" }, "Overrides": { "InstanceType": "c4.xlarge", "AvailabilityZone": "us-east-1a", } }, "Lifecycle": "on-demand", "ErrorCode": "InsufficientInstanceCapacity", "ErrorMessage": "" }, ], "Instances": [ { "LaunchTemplateAndOverrides": { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-01234a567b8910abcEXAMPLE", "Version": "1" }, "Overrides": { "InstanceType": "c5.large", "AvailabilityZone": "us-east-1a" } }, "Lifecycle": "on-demand", "InstanceIds": [ "i-1234567890abcdef0", "i-9876543210abcdef9" ] ] } ``` The following is example output for a fleet of type `instant` that launched no instances. ``` { "FleetId": "fleet-12a34b55-67cd-8ef9-ba9b-9208dEXAMPLE", "Errors": [ { "LaunchTemplateAndOverrides": { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-01234a567b8910abcEXAMPLE", "Version": "1" }, "Overrides": { "InstanceType": "c4.xlarge", "AvailabilityZone": "us-east-1a", } }, "Lifecycle": "on-demand", "ErrorCode": "InsufficientCapacity", "ErrorMessage": "" }, { "LaunchTemplateAndOverrides": { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-01234a567b8910abcEXAMPLE", "Version": "1" }, "Overrides": { "InstanceType": "c5.large", "AvailabilityZone": "us-east-1a", } }, "Lifecycle": "on-demand", "ErrorCode": "InsufficientCapacity", "ErrorMessage": "" }, ], "Instances": [] } ``` ## Create an EC2 Fleet that replaces unhealthy Spot Instances EC2 Fleet checks the health status of the instances in the fleet every two minutes. The health status of an instance is either `healthy` or `unhealthy`. EC2 Fleet determines the health status of an instance by using the status checks provided by Amazon EC2. An instance is determined as `unhealthy` when the status of either the instance status check or the system status check is `impaired` for three consecutive health status checks. For more information, see [Status checks for Amazon EC2 instances](monitoring-system-instance-status-check.md). You can configure your fleet to replace unhealthy Spot Instances. After setting `ReplaceUnhealthyInstances` to `true`, a Spot Instance is replaced when it is reported as `unhealthy`. The fleet can go below its target capacity for up to a few minutes while an unhealthy Spot Instance is being replaced. **Requirements** + Health check replacement is supported only for EC2 Fleets that maintain a target capacity (fleets of type `maintain`), and not for fleets of type `request` or `instant`. + Health check replacement is supported only for Spot Instances. This feature is not supported for On-Demand Instances. + You can configure your EC2 Fleet to replace unhealthy instances only when you create it. + Users can use health check replacement only if they have permission to call the `ec2:DescribeInstanceStatus` action. **To configure an EC2 Fleet to replace unhealthy Spot Instances** 1. Use the information for creating an EC2 Fleet in [Create an EC2 Fleet](#create-ec2-fleet-procedure). 1. To configure the fleet to replace unhealthy Spot Instances, in the JSON file, for `ReplaceUnhealthyInstances`, specify `true`. ## View all the EC2 Fleet configuration options To view the full list of EC2 Fleet configuration parameters, you can generate a JSON file. For a description of each parameter, see [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html). **To generate a JSON file with all possible EC2 Fleet parameters** Use the [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) (AWS CLI) command and the `--generate-cli-skeleton` parameter to generate an EC2 Fleet JSON file, and direct the output to a file to save it. ``` aws ec2 create-fleet \ --generate-cli-skeleton input > ec2createfleet.json ``` The following is example output. ``` { "DryRun": true, "ClientToken": "", "SpotOptions": { "AllocationStrategy": "price-capacity-optimized", "MaintenanceStrategies": { "CapacityRebalance": { "ReplacementStrategy": "launch" } }, "InstanceInterruptionBehavior": "hibernate", "InstancePoolsToUseCount": 0, "SingleInstanceType": true, "SingleAvailabilityZone": true, "MinTargetCapacity": 0, "MaxTotalPrice": "" }, "OnDemandOptions": { "AllocationStrategy": "prioritized", "CapacityReservationOptions": { "UsageStrategy": "use-capacity-reservations-first" }, "SingleInstanceType": true, "SingleAvailabilityZone": true, "MinTargetCapacity": 0, "MaxTotalPrice": "" }, "ExcessCapacityTerminationPolicy": "termination", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "", "LaunchTemplateName": "", "Version": "" }, "Overrides": [ { "InstanceType": "r5.metal", "MaxPrice": "", "SubnetId": "", "AvailabilityZone": "", "WeightedCapacity": 0.0, "Priority": 0.0, "Placement": { "AvailabilityZone": "", "Affinity": "", "GroupName": "", "PartitionNumber": 0, "HostId": "", "Tenancy": "dedicated", "SpreadDomain": "", "HostResourceGroupArn": "" }, "InstanceRequirements": { "VCpuCount": { "Min": 0, "Max": 0 }, "MemoryMiB": { "Min": 0, "Max": 0 }, "CpuManufacturers": [ "amd" ], "MemoryGiBPerVCpu": { "Min": 0.0, "Max": 0.0 }, "ExcludedInstanceTypes": [ "" ], "InstanceGenerations": [ "previous" ], "SpotMaxPricePercentageOverLowestPrice": 0, "OnDemandMaxPricePercentageOverLowestPrice": 0, "BareMetal": "included", "BurstablePerformance": "required", "RequireHibernateSupport": true, "NetworkInterfaceCount": { "Min": 0, "Max": 0 }, "LocalStorage": "excluded", "LocalStorageTypes": [ "ssd" ], "TotalLocalStorageGB": { "Min": 0.0, "Max": 0.0 }, "BaselineEbsBandwidthMbps": { "Min": 0, "Max": 0 }, "AcceleratorTypes": [ "inference" ], "AcceleratorCount": { "Min": 0, "Max": 0 }, "AcceleratorManufacturers": [ "amd" ], "AcceleratorNames": [ "a100" ], "AcceleratorTotalMemoryMiB": { "Min": 0, "Max": 0 } } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 0, "OnDemandTargetCapacity": 0, "SpotTargetCapacity": 0, "DefaultTargetCapacityType": "on-demand", "TargetCapacityUnitType": "memory-mib" }, "TerminateInstancesWithExpiration": true, "Type": "instant", "ValidFrom": "1970-01-01T00:00:00", "ValidUntil": "1970-01-01T00:00:00", "ReplaceUnhealthyInstances": true, "TagSpecifications": [ { "ResourceType": "fleet", "Tags": [ { "Key": "", "Value": "" } ] } ], "Context": "" } ``` # Tag a new or existing EC2 Fleet request and the instances and volumes it launches To help categorize and manage your EC2 Fleet requests and the instances and volumes that it launches, you can tag them with custom metadata. You can assign a tag to an EC2 Fleet request when you create it, or afterward. Similarly, you can assign a tag to the instances and volumes when they're launched by the fleet, or afterward. When you tag a fleet request, the instances and volumes that are launched by the fleet are not automatically tagged. You need to explicitly tag the instances and volumes launched by the fleet. You can choose to assign tags to only the fleet request, or to only the instances launched by the fleet, or to only the volumes attached to the instances launched by the fleet, or to all of them. **Note** For `instant` fleet types, you can tag volumes that are attached to On-Demand Instances and Spot Instances. For `request` or `maintain` fleet types, you can only tag volumes that are attached to On-Demand Instances. For more information about how tags work, see [Tag your Amazon EC2 resources](Using_Tags.md). **Prerequisite** Grant the user the permission to tag resources. For more information, see [Example: Tag resources](ExamplePolicies_EC2.md#iam-example-taggingresources). **To grant a user the permission to tag resources** Create a IAM policy that includes the following: + The `ec2:CreateTags` action. This grants the user permission to create tags. + The `ec2:CreateFleet` action. This grants the user permission to create an EC2 Fleet request. + For `Resource`, we recommend that you specify `"*"`. This allows users to tag all resource types. To provide access, add permissions to your users, groups, or roles: + Users and groups in AWS IAM Identity Center: Create a permission set. Follow the instructions in [Create a permission set](https://docs.aws.amazon.com//singlesignon/latest/userguide/howtocreatepermissionset.html) in the *AWS IAM Identity Center User Guide*. + Users managed in IAM through an identity provider: Create a role for identity federation. Follow the instructions in [Create a role for a third-party identity provider (federation)](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_roles_create_for-idp.html) in the *IAM User Guide*. + IAM users: + Create a role that your user can assume. Follow the instructions in [Create a role for an IAM user](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_roles_create_for-user.html) in the *IAM User Guide*. + (Not recommended) Attach a policy directly to a user or add a user to a user group. Follow the instructions in [Adding permissions to a user (console)](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_users_change-permissions.html#users_change_permissions-add-console) in the *IAM User Guide*. **To tag a new EC2 Fleet request** To tag an EC2 Fleet request when you create it, specify the key-value pair in the [JSON file](create-ec2-fleet.md#ec2-fleet-cli-skeleton) used to create the fleet. The value for `ResourceType` must be `fleet`. If you specify another value, the fleet request fails. **To tag instances and volumes launched by an EC2 Fleet** To tag instances and volumes when they are launched by the fleet, specify the tags in the [launch template](create-launch-template.md) that is referenced in the EC2 Fleet request. **Note** You can't tag volumes attached to Spot Instances that are launched by a `request` or `maintain` fleet type. **To tag an existing EC2 Fleet request, instance, and volume** Use the [create-tags](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-tags.html) command to tag existing resources. ``` aws ec2 create-tags \ --resources fleet-12a34b55-67cd-8ef9-ba9b-9208dEXAMPLE i-1234567890abcdef0 vol-1234567890EXAMPLE \ --tags Key=purpose,Value=test ``` # Describe an EC2 Fleet, its instances, and its events You can describe your EC2 Fleet configuration, the instances in your EC2 Fleet, and the event history of your EC2 Fleet. **Topics** + [ ## Describe your EC2 Fleet ](#describe-all-ec2-fleets) + [ ## Describe all instances in your EC2 Fleet ](#describe-instances-in-ec2-fleet) + [ ## Describe the event history for your EC2 Fleet ](#describe-ec2-fleet-event-history) ## Describe your EC2 Fleet ------ #### [ AWS CLI ] **To describe your EC2 Fleet** Use the [describe-fleets](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-fleets.html) command. ``` aws ec2 describe-fleets \ --fleet-ids fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE ``` The following is example output. ``` { "Fleets": [ { "ActivityStatus": "fulfilled", "CreateTime": "2022-02-09T03:35:52+00:00", "FleetId": "fleet-364457cd-3a7a-4ed9-83d0-7b63e51bb1b7", "FleetState": "active", "ExcessCapacityTerminationPolicy": "termination", "FulfilledCapacity": 2.0, "FulfilledOnDemandCapacity": 0.0, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "$Latest" } } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 2, "OnDemandTargetCapacity": 0, "SpotTargetCapacity": 2, "DefaultTargetCapacityType": "spot" }, "TerminateInstancesWithExpiration": false, "Type": "maintain", "ReplaceUnhealthyInstances": false, "SpotOptions": { "AllocationStrategy": "capacity-optimized", "InstanceInterruptionBehavior": "terminate" }, "OnDemandOptions": { "AllocationStrategy": "lowestPrice" } } ] } ``` ------ #### [ PowerShell ] **To describe your EC2 Fleet** Use the [Get-EC2FleetList](https://docs.aws.amazon.com/powershell/latest/reference/items/Get-EC2FleetList.html) cmdlet. ``` Get-EC2FleetList ` -FleetId fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE ``` ------ ## Describe all instances in your EC2 Fleet The returned list of running instances is refreshed periodically and might be out of date. ------ #### [ AWS CLI ] **To describe the instances for the specified EC2 Fleet** Use the [describe-fleet-instances](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-fleet-instances.html) command. ``` aws ec2 describe-fleet-instances \ --fleet-id fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE ``` The following is example output. ``` { "ActiveInstances": [ { "InstanceId": "i-09cd595998cb3765e", "InstanceHealth": "healthy", "InstanceType": "m4.large", "SpotInstanceRequestId": "sir-86k84j6p" }, { "InstanceId": "i-09cf95167ca219f17", "InstanceHealth": "healthy", "InstanceType": "m4.large", "SpotInstanceRequestId": "sir-dvxi7fsm" } ], "FleetId": "fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" } ``` ------ #### [ PowerShell ] **To describe the instances for the specified EC2 Fleet** Use the [Get-EC2FleetInstanceList](https://docs.aws.amazon.com/powershell/latest/reference/items/Get-EC2FleetInstanceList.html) cmdlet. ``` Get-EC2FleetInstanceList ` -FleetId fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE ``` ------ ## Describe the event history for your EC2 Fleet For more information about the events in the event history, see [EC2 Fleet event types](monitor-ec2-fleet-using-eventbridge.md#ec2-fleet-event-types). ------ #### [ AWS CLI ] **To describe the events for the specified EC2 Fleet** Use the [describe-fleet-history](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-fleet-history.html) command. ``` aws ec2 describe-fleet-history \ --fleet-id fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --start-time 2020-06-01T00:00:00Z ``` The following is example output. ``` { "HistoryRecords": [ { "EventInformation": { "EventSubType": "submitted" }, "EventType": "fleetRequestChange", "Timestamp": "2020-09-01T18:26:05.000Z" }, { "EventInformation": { "EventSubType": "active" }, "EventType": "fleetRequestChange", "Timestamp": "2020-09-01T18:26:15.000Z" }, { "EventInformation": { "EventDescription": "t2.small, ami-07c8bc5c1ce9598c3, ...", "EventSubType": "progress" }, "EventType": "fleetRequestChange", "Timestamp": "2020-09-01T18:26:17.000Z" }, { "EventInformation": { "EventDescription": "{\"instanceType\":\"t2.small\", ...}", "EventSubType": "launched", "InstanceId": "i-083a1c446e66085d2" }, "EventType": "instanceChange", "Timestamp": "2020-09-01T18:26:17.000Z" }, { "EventInformation": { "EventDescription": "{\"instanceType\":\"t2.small\", ...}", "EventSubType": "launched", "InstanceId": "i-090db02406cc3c2d6" }, "EventType": "instanceChange", "Timestamp": "2020-09-01T18:26:17.000Z" } ], "FleetId": "fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE", "LastEvaluatedTime": "1970-01-01T00:00:00.000Z", "StartTime": "2020-06-01T00:00:00.000Z" } ``` ------ #### [ PowerShell ] **To describe the events for the specified EC2 Fleet** Use the [Get-EC2FleetHistory](https://docs.aws.amazon.com/powershell/latest/reference/items/Get-EC2FleetHistory.html) cmdlet. ``` Get-EC2FleetHistory ` -FleetId fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE ` -UtcStartTime 2020-06-01T00:00:00Z ``` ------ # Modify an EC2 Fleet You can modify the total target capacity, Spot capacity, and On-Demand capacity of an EC2 Fleet. You can also modify whether running instances should be terminated if the new total target capacity is reduced below the current size of the fleet. ## Considerations Consider the following when modifying an EC2 Fleet: + **Fleet type** – You can only modify an EC2 Fleet of type `maintain`. You can't modify an EC2 Fleet of type `request` or `instant`. + **Fleet parameters** – You can modify the following parameters of an EC2 Fleet: + `target-capacity-specification` – Increase or decrease the target capacity for: + `TotalTargetCapacity` + `OnDemandTargetCapacity` + `SpotTargetCapacity` + `excess-capacity-termination-policy` – Whether running instances should be terminated if the total target capacity of the EC2 Fleet is decreased below the current size of the fleet. Valid values are: + `no-termination` + `termination` + **Fleet behavior when increasing total target capacity** – When you increase the total target capacity, the EC2 Fleet launches the additional instances according to the instance purchasing option specified for `DefaultTargetCapacityType`, which is either On-Demand Instances or Spot Instances, and according to the specified [allocation strategy](ec2-fleet-allocation-strategy.md). + **Fleet behavior when decreasing Spot target capacity** – When you decrease the Spot target capacity, the EC2 Fleet deletes any open requests that exceed the new target capacity. You can request that the fleet terminate Spot Instances until the size of the fleet reaches the new target capacity. When an EC2 Fleet terminates a Spot Instance because the target capacity was decreased, the instance receives a Spot Instance interruption notice. Instances are selected for termination based on the allocation strategy: + `capacity-optimized` – Terminates instances from pools with the least available capacity. + `price-capacity-optimized` – Uses a combination of price and available capacity: terminates instances from pools with the least available capacity and which are the highest-priced among these pools. + `diversified` – Terminates instances across all pools. + `lowest-price` – Terminates instances from highest-priced pools. Alternatively, you can request that EC2 Fleet keep the fleet at its current size, but not replace any Spot Instances that are interrupted or that you terminate manually. + **Fleet state** – You can modify an EC2 Fleet that is in the `submitted` or `active` state. When you modify a fleet, it enters the `modifying` state. ## Commands for modifying an EC2 Fleet ------ #### [ AWS CLI ] **To modify the total target capacity of an EC2 Fleet** Use the [modify-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/modify-fleet.html) command. ``` aws ec2 modify-fleet \ --fleet-id fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --target-capacity-specification TotalTargetCapacity=20 ``` If you are decreasing the target capacity but want to keep the fleet at its current size, you can modify the previous example as follows. ``` aws ec2 modify-fleet \ --fleet-id fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --target-capacity-specification TotalTargetCapacity=10 \ --excess-capacity-termination-policy no-termination ``` ------ #### [ PowerShell ] **To modify the total target capacity of an EC2 Fleet** Use the [Edit-EC2Fleet](https://docs.aws.amazon.com/powershell/latest/reference/items/Edit-EC2Fleet.html) cmdlet. ``` Edit-EC2Fleet ` -FleetId "fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -TargetCapacitySpecification_TotalTargetCapacity 20 ``` If you are decreasing the target capacity but want to keep the fleet at its current size, you can modify the previous example as follows. ``` Edit-EC2Fleet ` -FleetId "fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -TargetCapacitySpecification_TotalTargetCapacity 20 ` -ExcessCapacityTerminationPolicy "NoTermination" ``` ------ # Delete an EC2 Fleet request and the instances in the fleet If you no longer require an EC2 Fleet request, you can delete it. After you delete a fleet request, all Spot requests associated with the fleet are canceled, so that no new Spot Instances are launched. When you delete an EC2 Fleet request, you must also specify if you want to terminate all of its instances. These include both On-Demand Instances and Spot Instances. For `instant` fleets, EC2 Fleet must terminate the instances when the fleet is deleted. A deleted `instant` fleet with running instances is not supported. **Warning** **Terminating an instance is permanent and irreversible.** After you terminate an instance, you can no longer connect to it, and it can't be recovered. All attached Amazon EBS volumes that are configured to be deleted on termination are also permanently deleted and can't be recovered. All data stored on instance store volumes is permanently lost. For more information, see [How instance termination works](how-ec2-instance-termination-works.md). Before you terminate an instance, ensure that you have backed up all data that you need to retain after the termination to persistent storage. If you specify that the instances must be terminated when the fleet request is deleted, the fleet request enters the `deleted_terminating` state. Otherwise, it enters the `deleted_running` state, and the instances continue to run until they are interrupted or you terminate them manually. **Restrictions** + You can delete up to 25 fleets of type `instant` in a single operation. + You can delete up to 100 fleets of type `maintain` or `request` in a single operation. + You can delete up to 125 fleets in a single operation, provided you do not exceed the quota for each fleet type, as specified above. + If you exceed the specified number of fleets to delete, no fleets are deleted. + A deleted `instant` fleet with running instances is not supported. When you delete an `instant` fleet, Amazon EC2 automatically terminates all its instances. For `instant` fleets with more than 1000 instances, the deletion request might fail. If your fleet has more than 1000 instances, first terminate most of the instances manually, leaving 1000 or fewer. Then delete the fleet, and the remaining instances will be terminated automatically. ------ #### [ AWS CLI ] **To delete an EC2 Fleet request and terminate its instances** Use the [delete-fleets](https://docs.aws.amazon.com/cli/latest/reference/ec2/delete-fleets.html) command with the `--terminate-instances` option. ``` aws ec2 delete-fleets \ --fleet-ids fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --terminate-instances ``` The following is example output. ``` { "UnsuccessfulFleetDeletions": [], "SuccessfulFleetDeletions": [ { "CurrentFleetState": "deleted_terminating", "PreviousFleetState": "active", "FleetId": "fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" } ] } ``` **To delete an EC2 Fleet request without terminating its instances** Modify the previous example by using the `--no-terminate-instances` option instead. Note that `--no-terminate-instances` is not supported for `instant` fleets. ``` aws ec2 delete-fleets \ --fleet-ids fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --no-terminate-instances ``` The following is example output. ``` { "UnsuccessfulFleetDeletions": [], "SuccessfulFleetDeletions": [ { "CurrentFleetState": "deleted_running", "PreviousFleetState": "active", "FleetId": "fleet-4b8aaae8-dfb5-436d-a4c6-3dafa4c6b7dcEXAMPLE" } ] } ``` ------ #### [ PowerShell ] **To delete an EC2 Fleet request and terminate its instances** Use the [Remove-EC2Fleet](https://docs.aws.amazon.com/powershell/latest/reference/items/Remove-EC2Fleet.html) cmdlet with the `-TerminateInstance` parameter. ``` Remove-EC2Fleet ` -FleetId "fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -TerminateInstance $true ``` **To delete an EC2 Fleet request without terminating its instances** Modify the previous example by changing the value of the `-TerminateInstance` parameter. ``` Remove-EC2Fleet ` -FleetId "fleet-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -TerminateInstance $false ``` ------ ## Troubleshoot when a fleet fails to delete If an EC2 Fleet request fails to delete, `UnsuccessfulFleetDeletions` in the output returns the ID of the EC2 Fleet request, an error code, and an error message. The error codes are: + `ExceededInstantFleetNumForDeletion` + `fleetIdDoesNotExist` + `fleetIdMalformed` + `fleetNotInDeletableState` + `NoTerminateInstancesNotSupported` + `UnauthorizedOperation` + `unexpectedError` **Troubleshoot `ExceededInstantFleetNumForDeletion`** If you try to delete more than 25 `instant` fleets in a single request, the `ExceededInstantFleetNumForDeletion` error is returned. The following is example output for this error. ``` { "UnsuccessfulFleetDeletions": [ { "FleetId": " fleet-5d130460-0c26-bfd9-2c32-0100a098f625", "Error": { "Message": "Can’t delete more than 25 instant fleets in a single request.", "Code": "ExceededInstantFleetNumForDeletion" } }, { "FleetId": "fleet-9a941b23-0286-5bf4-2430-03a029a07e31", "Error": { "Message": "Can’t delete more than 25 instant fleets in a single request.", "Code": "ExceededInstantFleetNumForDeletion" } } . . . ], "SuccessfulFleetDeletions": [] } ``` **Troubleshoot `NoTerminateInstancesNotSupported`** If you specify that the instances in an `instant` fleet must not be terminated when you delete the fleet, the `NoTerminateInstancesNotSupported` error is returned. `--no-terminate-instances` is not supported for `instant` fleets. The following is example output for this error. ``` { "UnsuccessfulFleetDeletions": [ { "FleetId": "fleet-5d130460-0c26-bfd9-2c32-0100a098f625", "Error": { "Message": "NoTerminateInstances option is not supported for instant fleet", "Code": "NoTerminateInstancesNotSupported" } } ], "SuccessfulFleetDeletions": [] } ``` **Troubleshoot `UnauthorizedOperation`** If you do not have permission to terminate instances, you get the `UnauthorizedOperation` error when deleting a fleet that must terminate its instances. The following is the error response. ``` UnauthorizedOperationYou are not authorized to perform this operation. Encoded authorization failure message: VvuncIxj7Z_CPGNYXWqnuFV-YjByeAU66Q9752NtQ-I3-qnDLWs6JLFd KnSMMiq5s6cGqjjPtEDpsnGHzzyHasFHOaRYJpaDVravoW25azn6KNkUQQlFwhJyujt2dtNCdduJfrqcFYAjlEiRMkfDHt7N63SKlweKUl BHturzDK6A560Y2nDSUiMmAB1y9UNtqaZJ9SNe5sNxKMqZaqKtjRbk02RZu5V2vn9VMk6fm2aMVHbY9JhLvGypLcMUjtJ76H9ytg2zRlje VPiU5v2s-UgZ7h0p2yth6ysUdhlONg6dBYu8_y_HtEI54invCj4CoK0qawqzMNe6rcmCQHvtCxtXsbkgyaEbcwmrm2m01-EMhekLFZeJLr DtYOpYcEl4_nWFX1wtQDCnNNCmxnJZAoJvb3VMDYpDTsxjQv1PxODZuqWHs23YXWVywzgnLtHeRf2o4lUhGBw17mXsS07k7XAfdPMP_brO PT9vrHtQiILor5VVTsjSPWg7edj__1rsnXhwPSu8gI48ZLRGrPQqFq0RmKO_QIE8N8s6NWzCK4yoX-9gDcheurOGpkprPIC9YPGMLK9tug 89b1215c-7814-40ae-a8db-41761f43f2b0 ``` To resolve the error, you must add the `ec2:TerminateInstances` action to the IAM policy, as shown in the following example. # Work with Spot Fleet To start using a Spot Fleet, create a request that includes the total target capacity for Spot Instances, an optional On-Demand portion, and either manually specify an AMI and a key pair, or specify a launch template that includes the configuration for the instances in the fleet. You can optionally specify additional parameters, or let the fleet use default values. You can also tag the fleet request, and its instances and volumes, when you create the fleet. The fleet launches On-Demand Instances when there is available capacity, and launches Spot Instances when your maximum price exceeds the Spot price and capacity is available. Once your fleet is launched, you can describe the fleet request, the instances in the fleet, and any fleet events. You can also assign additional tags as needed. If you need to change any fleet parameters, such as the total target capacity, you can modify the fleet, provided it was configured to maintain capacity. You can't modify the capacity of a one-time request after it's been submitted. The fleet request remains active until it expires or you cancel (delete) it. When you cancel the fleet request, you can either terminate the instances or leave them running. If you choose to leave them running, the On-Demand Instances run until you terminate them, and the Spot Instances run until they're interrupted or you terminate them. **Topics** + [ # Spot Fleet request states ](spot-fleet-states.md) + [ # Spot Fleet permissions ](spot-fleet-prerequisites.md) + [ # Create a Spot Fleet ](create-spot-fleet.md) + [ # Tag a new or existing Spot Fleet request and the instances and volumes it launches ](tag-spot-fleet.md) + [ # Describe a Spot Fleet request, its instances, and event history ](manage-spot-fleet.md) + [ # Modify a Spot Fleet request ](modify-spot-fleet.md) + [ # Cancel (delete) a Spot Fleet request ](cancel-spot-fleet.md) + [ # Understand automatic scaling for Spot Fleet ](spot-fleet-automatic-scaling.md) # Spot Fleet request states A Spot Fleet request can be one of various states, with each state indicating a different stage of the request's lifecycle and its management of instances. A Spot Fleet request can be in one of the following states: `submitted` The Spot Fleet request is being evaluated and Amazon EC2 is preparing to launch the target number of instances. If a request would exceed your Spot Fleet quotas, it is canceled immediately. `active` The Spot Fleet has been validated and Amazon EC2 is attempting to maintain the target number of running Spot Instances. The request remains in this state until it is modified or canceled. `modifying` The Spot Fleet request is being modified. The request remains in this state until the modification is fully processed or the request is canceled. Only a `maintain` fleet type can be modified. This state does not apply to a one-time `request` fleet type. `cancelled_running` The Spot Fleet is canceled (deleted) and does not launch additional Spot Instances. Its existing instances continue to run until they are interrupted or terminated manually. The request remains in this state until all instances are interrupted or terminated. `cancelled_terminating` The Spot Fleet is canceled (deleted) and its instances are terminating. The request remains in this state until all instances are terminated. `cancelled` The Spot Fleet is canceled (deleted) and has no running instances. The request is deleted two days after its instances are terminated. # Spot Fleet permissions If your users will create or manage a Spot Fleet, you need to grant them the required permissions. If you use the Amazon EC2 console to create a Spot Fleet, it creates two service-linked roles named `AWSServiceRoleForEC2SpotFleet` and `AWSServiceRoleForEC2Spot`, and a role named `aws-ec2-spot-fleet-tagging-role` that grant the Spot Fleet the permissions to request, launch, terminate, and tag resources on your behalf. If you use the AWS CLI or an API, you must ensure that these roles exist. Use the following instructions to grant the required permissions and create the roles. **Topics** + [ ## Grant permission to users for Spot Fleet ](#spot-fleet-iam-users) + [ ## Service-linked role for Spot Fleet ](#service-linked-roles-spot-fleet-requests) + [ ## Service-linked role for Spot Instances ](#service-linked-roles-spot-instances) + [ ## IAM role for tagging a Spot Fleet ](#spot-fleet-service-linked-role) ## Grant permission to users for Spot Fleet If your users will create or manage a Spot Fleet, be sure to grant them the required permissions. **To create a policy for Spot Fleet** 1. Open the IAM console at [https://console.aws.amazon.com/iam/](https://console.aws.amazon.com/iam/). 1. In the navigation pane, choose **Policies**, **Create policy**. 1. On the **Create policy** page, choose **JSON**, and replace the text with the following. ------ #### [ JSON ] **** ``` { "Version":"2012-10-17", "Statement": [ { "Effect": "Allow", "Action": [ "ec2:RunInstances", "ec2:CreateTags", "ec2:RequestSpotFleet", "ec2:ModifySpotFleetRequest", "ec2:CancelSpotFleetRequests", "ec2:DescribeSpotFleetRequests", "ec2:DescribeSpotFleetInstances", "ec2:DescribeSpotFleetRequestHistory" ], "Resource": "*" }, { "Effect": "Allow", "Action": "iam:PassRole", "Resource": "arn:aws:iam::*:role/aws-ec2-spot-fleet-tagging-role" }, { "Effect": "Allow", "Action": [ "iam:CreateServiceLinkedRole", "iam:ListRoles", "iam:ListInstanceProfiles" ], "Resource": "*" } ] } ``` ------ The preceding example policy grants a user the permissions required for most Spot Fleet use cases. To limit the user to specific API actions, specify only those API actions instead. **Required EC2 and IAM APIs** The following APIs must be included in the policy: + `ec2:RunInstances` – Required to launch instances in a Spot Fleet + `ec2:CreateTags` – Required to tag the Spot Fleet request, instances, or volumes + `iam:PassRole` – Required to specify the Spot Fleet role + `iam:CreateServiceLinkedRole` – Required to create the service-linked role + `iam:ListRoles` – Required to enumerate existing IAM roles + `iam:ListInstanceProfiles` – Required to enumerate existing instance profiles **Important** If you specify a role for the IAM instance profile in the launch specification or launch template, you must grant the user the permission to pass the role to the service. To do this, in the IAM policy include `"arn:aws:iam::*:role/IamInstanceProfile-role"` as a resource for the `iam:PassRole` action. For more information, see [Granting a user permissions to pass a role to an AWS service](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_use_passrole.html) in the *IAM User Guide*. **Spot Fleet APIs** Add the following Spot Fleet API actions to your policy, as needed: + `ec2:RequestSpotFleet` + `ec2:ModifySpotFleetRequest` + `ec2:CancelSpotFleetRequests` + `ec2:DescribeSpotFleetRequests` + `ec2:DescribeSpotFleetInstances` + `ec2:DescribeSpotFleetRequestHistory` **Optional IAM APIs** (Optional) To enable a user to create roles or instance profiles using the IAM console, you must add the following actions to the policy: + `iam:AddRoleToInstanceProfile` + `iam:AttachRolePolicy` + `iam:CreateInstanceProfile` + `iam:CreateRole` + `iam:GetRole` + `iam:ListPolicies` 1. Choose **Review policy**. 1. On the **Review policy** page, enter a policy name and description, and choose **Create policy**. 1. To provide access, add permissions to your users, groups, or roles: + Users and groups in AWS IAM Identity Center: Create a permission set. Follow the instructions in [Create a permission set](https://docs.aws.amazon.com//singlesignon/latest/userguide/howtocreatepermissionset.html) in the *AWS IAM Identity Center User Guide*. + Users managed in IAM through an identity provider: Create a role for identity federation. Follow the instructions in [Create a role for a third-party identity provider (federation)](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_roles_create_for-idp.html) in the *IAM User Guide*. + IAM users: + Create a role that your user can assume. Follow the instructions in [Create a role for an IAM user](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_roles_create_for-user.html) in the *IAM User Guide*. + (Not recommended) Attach a policy directly to a user or add a user to a user group. Follow the instructions in [Adding permissions to a user (console)](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_users_change-permissions.html#users_change_permissions-add-console) in the *IAM User Guide*. ## Service-linked role for Spot Fleet Amazon EC2 uses service-linked roles for the permissions that it requires to call other AWS services on your behalf. A service-linked role is a unique type of IAM role that is linked directly to an AWS service. Service-linked roles provide a secure way to delegate permissions to AWS services because only the linked service can assume a service-linked role. For more information, see [Service-linked roles](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_create-service-linked-role.html) in the *IAM User Guide*. Amazon EC2 uses the service-linked role named **AWSServiceRoleForEC2SpotFleet** to launch and manage instances on your behalf. **Important** If you specify an [encrypted AMI](AMIEncryption.md) or an encrypted Amazon EBS snapshot in your Spot Fleet, you must grant the **AWSServiceRoleForEC2SpotFleet** role permission to use the CMK so that Amazon EC2 can launch instances on your behalf. For more information, see [Grant access to CMKs for use with encrypted AMIs and EBS snapshots](#spot-fleet-service-linked-roles-access-to-cmks). ### Permissions granted by AWSServiceRoleForEC2SpotFleet The **AWSServiceRoleForEC2SpotFleet** role grants the Spot Fleet permission to request, launch, terminate, and tag instances on your behalf. Amazon EC2 uses this service-linked role to complete the following actions: + `ec2:RequestSpotInstances` - Request Spot Instances + `ec2:RunInstances` - Launch instances + `ec2:TerminateInstances` - Terminate instances + `ec2:DescribeImages` - Describe Amazon Machine Images (AMIs) for the instances + `ec2:DescribeInstanceStatus` - Describe the status of the instances + `ec2:DescribeSubnets` - Describe the subnets for the instances + `ec2:CreateTags` - Add tags to the Spot Fleet request, instances, and volumes + `elasticloadbalancing:RegisterInstancesWithLoadBalancer` - Add the specified instances to the specified load balancer + `elasticloadbalancing:RegisterTargets` - Register the specified targets with the specified target group ### Create the service-linked role Under most circumstances, you don't need to manually create a service-linked role. Amazon EC2 creates the **AWSServiceRoleForEC2SpotFleet** service-linked role the first time you create a Spot Fleet using the console. If you had an active Spot Fleet request before October 2017, when Amazon EC2 began supporting this service-linked role, Amazon EC2 created the **AWSServiceRoleForEC2SpotFleet** role in your AWS account. For more information, see [A new role appeared in my AWS account](https://docs.aws.amazon.com/IAM/latest/UserGuide/troubleshoot_roles.html#troubleshoot_roles_new-role-appeared) in the *IAM User Guide*. If you use the AWS CLI or an API to create a Spot Fleet, you must first ensure that this role exists. **To create the AWSServiceRoleForEC2SpotFleet role for Spot Fleet using the console** 1. Open the IAM console at [https://console.aws.amazon.com/iam/](https://console.aws.amazon.com/iam/). 1. In the navigation pane, choose **Roles**. 1. Choose **Create role**. 1. On the **Select trusted entity** page, do the following: 1. For **Trusted entity type**, choose **AWS service**. 1. Under **Use case**, for **Service or use case**, choose **EC2**. 1. For **Use case**, choose **EC2 - Spot Fleet**. **Note** The **EC2 - Spot Fleet** use case will automatically create a policy with the required IAM permissions and will suggest **AWSEC2SpotFleetServiceRolePolicy** as the role name. 1. Choose **Next**. 1. On the **Add permissions** page, choose **Next**. 1. On the **Name, review, and create** page, choose **Create role**. **To create the AWSServiceRoleForEC2SpotFleet role for Spot Fleet using the AWS CLI** Use the [create-service-linked-role](https://docs.aws.amazon.com/cli/latest/reference/iam/create-service-linked-role.html) command as follows. ``` aws iam create-service-linked-role --aws-service-name spotfleet.amazonaws.com ``` If you no longer need to use Spot Fleet, we recommend that you delete the **AWSServiceRoleForEC2SpotFleet** role. After this role is deleted from your account, Amazon EC2 will create the role again if you request a Spot Fleet using the console. For more information, see [Deleting a service-linked role](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_manage_delete.html#id_roles_manage_delete_slr) in the *IAM User Guide*. ### Grant access to CMKs for use with encrypted AMIs and EBS snapshots If you specify an [encrypted AMI](AMIEncryption.md) or an encrypted Amazon EBS snapshot in your Spot Fleet request and you use a customer managed key for encryption, you must grant the **AWSServiceRoleForEC2SpotFleet** role permission to use the CMK so that Amazon EC2 can launch instances on your behalf. To do this, you must add a grant to the CMK, as shown in the following procedure. When providing permissions, grants are an alternative to key policies. For more information, see [Using Grants](https://docs.aws.amazon.com/kms/latest/developerguide/grants.html) and [Using Key Policies in AWS KMS](https://docs.aws.amazon.com/kms/latest/developerguide/key-policies.html) in the *AWS Key Management Service Developer Guide*. **To grant the AWSServiceRoleForEC2SpotFleet role permissions to use the CMK** + Use the [create-grant](https://docs.aws.amazon.com/cli/latest/reference/kms/create-grant.html) command to add a grant to the CMK and to specify the principal (the **AWSServiceRoleForEC2SpotFleet** service-linked role) that is given permission to perform the operations that the grant permits. The CMK is specified by the `key-id` parameter and the ARN of the CMK. The principal is specified by the `grantee-principal` parameter and the ARN of the **AWSServiceRoleForEC2SpotFleet** service-linked role. ``` aws kms create-grant \ --region us-east-1 \ --key-id arn:aws:kms:us-east-1:444455556666:key/1234abcd-12ab-34cd-56ef-1234567890ab \ --grantee-principal arn:aws:iam::111122223333:role/aws-service-role/spotfleet.amazonaws.com/AWSServiceRoleForEC2SpotFleet \ --operations "Decrypt" "Encrypt" "GenerateDataKey" "GenerateDataKeyWithoutPlaintext" "CreateGrant" "DescribeKey" "ReEncryptFrom" "ReEncryptTo" ``` ## Service-linked role for Spot Instances Amazon EC2 uses the service-linked role named **AWSServiceRoleForEC2Spot** to launch and manage Spot Instances on your behalf. For more information, see [Service-linked role for Spot Instance requests](service-linked-roles-spot-instance-requests.md). ## IAM role for tagging a Spot Fleet The `aws-ec2-spot-fleet-tagging-role` IAM role grants the Spot Fleet permission to tag the Spot Fleet request, instances, and volumes. For more information, see [Tag a new or existing Spot Fleet request and the instances and volumes it launches](tag-spot-fleet.md). **Important** If you choose to tag instances in the fleet and you also choose to maintain target capacity (the Spot Fleet request is of type `maintain`), the differences in the permissions that are set for the user and the `IamFleetRole` might lead to inconsistent tagging behavior of instances in the fleet. If the `IamFleetRole` does not include the `CreateTags` permission, some of the instances launched by the fleet might not be tagged. While we are working to fix this inconsistency, to ensure that all instances launched by the fleet are tagged, we recommend that you use the `aws-ec2-spot-fleet-tagging-role` role for the `IamFleetRole`. Alternatively, to use an existing role, attach the `AmazonEC2SpotFleetTaggingRole` AWS Managed Policy to the existing role. Otherwise, you need to manually add the `CreateTags` permission to your existing policy. **To create the IAM role for tagging a Spot Fleet** 1. Open the IAM console at [https://console.aws.amazon.com/iam/](https://console.aws.amazon.com/iam/). 1. In the navigation pane, choose **Roles**. 1. Choose **Create role**. 1. On the **Select trusted entity** page, under **Trusted entity type**, choose **AWS service**. 1. Under **Use case**, from **Use cases for other AWS services**, choose **EC2**, and then choose **EC2 - Spot Fleet Tagging**. 1. Choose **Next**. 1. On the **Add permissions** page, choose **Next**. 1. On the **Name, review, and create** page, for **Role name**, enter a name for the role (for example, **aws-ec2-spot-fleet-tagging-role**). 1. Review the information on the page, and then choose **Create role**. ### Cross-service confused deputy prevention The [confused deputy problem](https://docs.aws.amazon.com/IAM/latest/UserGuide/confused-deputy.html) is a security issue where an entity that doesn't have permission to perform an action can coerce a more-privileged entity to perform the action. We recommend that you use the [https://docs.aws.amazon.com/IAM/latest/UserGuide/reference_policies_condition-keys.html#condition-keys-sourcearn](https://docs.aws.amazon.com/IAM/latest/UserGuide/reference_policies_condition-keys.html#condition-keys-sourcearn) and [https://docs.aws.amazon.com/IAM/latest/UserGuide/reference_policies_condition-keys.html#condition-keys-sourceaccount](https://docs.aws.amazon.com/IAM/latest/UserGuide/reference_policies_condition-keys.html#condition-keys-sourceaccount) global condition context keys in the `aws-ec2-spot-fleet-tagging-role` trust policy to limit the permissions that Spot Fleet gives another service to the resource. **To add the aws:SourceArn and aws:SourceAccount condition keys to the `aws-ec2-spot-fleet-tagging-role` trust policy** 1. Open the IAM console at [https://console.aws.amazon.com/iam/](https://console.aws.amazon.com/iam/). 1. In the navigation pane, choose **Roles**. 1. Find the `aws-ec2-spot-fleet-tagging-role` that you created previously and choose the link (not the checkbox). 1. Under **Summary**, choose the **Trust relationships** tab, and then choose **Edit trust policy**. 1. In the JSON statement, add a `Condition` element containing your `aws:SourceAccount` and `aws:SourceArn` global condition context keys to prevent the [confused deputy problem](https://docs.aws.amazon.com/IAM/latest/UserGuide/confused-deputy.html), as follows: ``` "Condition": { "ArnLike": { "aws:SourceArn": "arn:aws:ec2:us-east-1:111122223333:spot-fleet-request/sfr-*" }, "StringEquals": { "aws:SourceAccount": "111122223333" } ``` **Note** If the `aws:SourceArn` value contains the account ID and you use both global condition context keys, the `aws:SourceAccount` value and the account in the `aws:SourceArn` value must use the same account ID when used in the same policy statement. The final trust policy will be as follows: ------ #### [ JSON ] **** ``` { "Version":"2012-10-17", "Statement": { "Sid": "ConfusedDeputyPreventionExamplePolicy", "Effect": "Allow", "Principal": { "Service": "spotfleet.amazonaws.com" }, "Action": "sts:AssumeRole", "Condition": { "ArnLike": { "aws:SourceArn": "arn:aws:ec2:us-east-1:111122223333:spot-fleet-request/sfr-*" }, "StringEquals": { "aws:SourceAccount": "111122223333" } } } } ``` ------ 1. Choose **Update policy**. The following table provides potential values for `aws:SourceArn` to limit the scope of the your `aws-ec2-spot-fleet-tagging-role` in varying degrees of specificity. **** | API operation | Called service | Scope | `aws:SourceArn` | | --- | --- | --- | --- | | RequestSpotFleet | AWS STS (AssumeRole) | Limit the AssumeRole capability on aws-ec2-spot-fleet-tagging-role to spot-fleet-requests in the specified account. | arn:aws:ec2:\$1:123456789012:spot-fleet-request/sfr-\$1 | | RequestSpotFleet | AWS STS (AssumeRole) | Limit the AssumeRole capability on aws-ec2-spot-fleet-tagging-role to spot-fleet-requests in the specified account and specified Region. Note that this role will not be usable in other Regions. | arn:aws:ec2:us-east-1:123456789012:spot-fleet-request/sfr-\$1 | | RequestSpotFleet | AWS STS (AssumeRole) | Limit the AssumeRole capability on aws-ec2-spot-fleet-tagging-role to only actions affecting the fleet sfr-11111111-1111-1111-1111-111111111111. Note that this role may not be usable for other Spot Fleets. Also, this role cannot be used to launch any new Spot Fleets through request-spot-fleet. | arn:aws:ec2:us-east-1:123456789012:spot-fleet-request/sfr-11111111-1111-1111-1111-111111111111 | # Create a Spot Fleet Using the AWS Management Console, you can quickly create a Spot Fleet request by choosing only an AMI and your desired total target capacity. Amazon EC2 will configure a fleet that best meets your needs and follows Spot best practices. Alternatively, you can modify any of the default settings. If you want to include On-Demand Instances in your fleet, you must specify a launch template in your request and specify you desired On-Demand capacity. The fleet launches On-Demand Instances when capacity is available, and launches Spot Instances when your maximum price exceeds the Spot price and capacity is available. If your fleet includes Spot Instances and is of type `maintain`, Amazon EC2 will attempt to maintain your fleet target capacity when your Spot Instances are interrupted. **Required permissions** For more information, see [Spot Fleet permissions](spot-fleet-prerequisites.md). **Topics** + [ ## Quickly create a Spot Fleet request ](#create-spot-fleet-quick) + [ ## Create a Spot Fleet request using defined parameters ](#create-spot-fleet-advanced) + [ ## Create a Spot Fleet that replaces unhealthy Spot Instances ](#spot-fleet-health-checks) ## Quickly create a Spot Fleet request Follow these steps to quickly create a Spot Fleet request using the Amazon EC2 console. **To create a Spot Fleet request using the recommended settings** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. If you are new to Spot, you see a welcome page; choose **Get started**. Otherwise, choose **Create Spot Fleet Request**. 1. Under **Launch parameters**, choose **Manually configure launch parameters**. 1. For **AMI**, choose an AMI. 1. Under **Target capacity**, for **Total target capacity**, specify the number of units to request. For the type of unit, you can choose **Instances**, **vCPUs**, or **Memory (GiB)**. 1. Under **Your fleet request at a glance**, review your fleet configuration, and choose **Launch**. ## Create a Spot Fleet request using defined parameters You can create a Spot Fleet by using parameters that you define. ------ #### [ Console ] **To create a Spot Fleet request using defined parameters** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. If you are new to Spot, you see a welcome page; choose **Get started**. Otherwise, choose **Create Spot Fleet Request**. 1. For **Launch parameters**, you can either manually configure the launch parameters or you can use a launch template, as follows: 1. [Manually configure] To define the launch parameters in the Amazon EC2 console, choose **Manually configure launch parameters**, and then do the following: 1. For **AMI**, choose one of the basic AMIs provided by AWS, or choose **Search for AMI** to use an AMI from our user community, the AWS Marketplace, or one of your own. **Note** If an AMI specified in the launch parameters is deregistered or disabled, no new instances can be launched from the AMI. For fleets that are set to maintain target capacity, the target capacity will not be maintained. 1. (Optional) For **Key pair name**, choose an existing key pair or create a new one. [Existing key pair] Choose the key pair. [New key pair] Choose **Create new key pair** to go the **Key pairs** page. When you are done, return to the **Spot Requests** page and refresh the list. 1. (Optional) Expand **Additional launch parameters**, and do the following: 1. (Optional) To enable Amazon EBS optimization, for **EBS-optimized**, select **Launch EBS-optimized instances**. 1. (Optional) To add temporary block-level storage for your instances, for **Instance store**, choose **Attach at launch**. 1. (Optional) To add storage, choose **Add new volume**, and specify additional instance store volumes or Amazon EBS volumes, depending on the instance type. 1. (Optional) By default, basic monitoring is enabled for your instances. To enable detailed monitoring, for **Monitoring**, select **Enable CloudWatch detailed monitoring**. 1. (Optional) To run a Dedicated Spot Instance, for **Tenancy**, choose **Dedicated - run a dedicated instance**. 1. (Optional) For **Security groups**, choose one or more security groups or create a new one. [Existing security group] Choose one or more security groups. [New security group] Choose **Create new security group** to go the **Security Groups** page. When you are done, return to the **Spot Requests** and refresh the list. 1. (Optional) To make your instances reachable from the internet, for **Auto-assign IPv4 Public IP**, choose **Enable**. 1. (Optional) To launch your Spot Instances with an IAM role, for **IAM instance profile**, choose the role. 1. (Optional) To run a start-up script, copy it to **User data**. 1. (Optional) To add a tag, choose **Create tag** and enter the key and value for the tag, and choose **Create**. Repeat for each tag. For each tag, to tag the instances and the Spot Fleet request with the same tag, ensure that both **Instances** and **Fleet** are selected. To tag only the instances launched by the fleet, clear **Fleet**. To tag only the Spot Fleet request, clear **Instances**. 1. [Launch template] To use a configuration you created in a launch template, choose **Use a launch template**, and for **Launch template**, choose a launch template. **Note** If you want On-Demand capacity in your Spot Fleet, you must specify a launch template. 1. For **Additional request details**, do the following: 1. Review the additional request details. To make changes, clear **Apply defaults**. 1. (Optional) For **IAM fleet role**, you can use the default role or choose a different role. To use the default role after changing the role, choose **Use default role**. 1. (Optional) To create a request that is valid only during a specific time period, edit **Request valid from** and **Request valid until**. 1. (Optional) By default, Amazon EC2 terminates your Spot Instances when the Spot Fleet request expires. To keep them running after your request expires, clear **Terminate the instances when the request expires**. 1. (Optional) To register your Spot Instances with a load balancer, choose **Receive traffic from one or more load balancers** and choose one or more Classic Load Balancers or target groups. 1. For **Target capacity**, do the following: 1. For **Total target capacity**, specify the number of units to request. For the type of unit, you can choose **Instances**, **vCPUs**, or **Memory (MiB)**. To specify a target capacity of 0 so that you can add capacity later, you must first select **Maintain target capacity**. 1. (Optional) For **Include On-Demand base capacity**, specify the number of On-Demand units to request. The number must be less than the **Total target capacity**. Amazon EC2 calculates the difference, and allocates the difference to Spot units to request. **Important** To specify optional On-Demand capacity, you must first choose a launch template. 1. (Optional) By default, Amazon EC2 terminates Spot Instances when they are interrupted. To maintain the target capacity, select **Maintain target capacity**. You can then specify that Amazon EC2 terminates, stops, or hibernates Spot Instances when they are interrupted. To do so, choose the corresponding option from **Interruption behavior**. **Note** If an AMI specified in the launch parameters is deregistered or disabled, no new instances can be launched from the AMI. In this case, for fleets that are set to maintain target capacity, the target capacity will not be maintained. 1. (Optional) To allow Spot Fleet to launch a replacement Spot Instance when an instance rebalance notification is emitted for an existing Spot Instance in the fleet, select **Capacity rebalance**, and then choose an instance replacement strategy. If you choose **Launch before terminate**, specify the delay (in seconds) before Amazon EC2 terminates the old instances. For more information, see [Use Capacity Rebalancing in EC2 Fleet and Spot Fleet to replace at-risk Spot Instances](ec2-fleet-capacity-rebalance.md). 1. (Optional) To control the amount you pay per hour for all the Spot Instances in your fleet, select **Set maximum cost for Spot Instances** and then enter the maximum total amount you're willing to pay per hour. When the maximum total amount is reached, Spot Fleet stops launching Spot Instances even if it hasn’t met the target capacity. For more information, see [Set a spending limit for your EC2 Fleet or Spot Fleet](ec2-fleet-control-spending.md). 1. For **Network**, do the following: 1. For **Network**, choose an existing VPC or create a new one. [Existing VPC] Choose the VPC. [New VPC] Choose **Create new VPC** to go the Amazon VPC console. When you're done, return to this screen and refresh the list. 1. (Optional) For **Availability Zone**, let Amazon EC2 choose the Availability Zones for your Spot Instances, or specify one or more Availability Zones. If you have more than one subnet in an Availability Zone, choose the appropriate subnet from **Subnet**. To add subnets, choose **Create new subnet** to go to the Amazon VPC console. When you are done, return to this screen and refresh the list. 1. For **Instance type requirements**, you can either specify instance attributes and let Amazon EC2 identify the optimal instance types with these attributes, or you can specify a list of instances. For more information, see [Specify attributes for instance type selection for EC2 Fleet or Spot Fleet](ec2-fleet-attribute-based-instance-type-selection.md). 1. If you choose **Specify instance attributes that match your compute requirements**, specify your instance attributes as follows: 1. For **vCPUs**, enter the desired minimum and maximum number of vCPUs. To specify no limit, select **No minimum** or **No maximum**, or both. 1. For **Memory (GiB)**, enter the desired minimum and maximum amount of memory. To specify no limit, select **No minimum** or **No maximum**, or both. 1. (Optional) For **Additional instance attribute**, you can optionally specify one or more attributes to express your compute requirements in more detail. Each additional attribute adds a further constraint to your request. You can omit the additional attributes; when omitted, the default values are used. For a description of each attribute and their default values, see [get-spot-placement-scores](https://docs.aws.amazon.com/cli/latest/reference/ec2/get-spot-placement-scores.html). 1. (Optional) To view the instance types with your specified attributes, expand **Preview matching instance types**. To exclude instance types from being used in your request, select the instances and then choose **Exclude selected instance types**. 1. If you choose **Manually select instance types**, Spot Fleet provides a default list of instance types. To select more instance types, choose **Add instance types**, select the instance types to use in your request, and choose **Select**. To delete instance types, select the instance types and choose **Delete**. 1. For **Allocation strategy**, choose a Spot allocation strategy and an On-Demand allocation strategy that meets your needs. For more information, see [Use allocation strategies to determine how EC2 Fleet or Spot Fleet fulfills Spot and On-Demand capacity](ec2-fleet-allocation-strategy.md). 1. For **Your fleet request at a glance**, review your fleet configuration, and make any adjustments if necessary. 1. (Optional) To download a copy of the launch configuration for use with the AWS CLI, choose **JSON config**. 1. When you're ready to launch your Spot Fleet, choose **Launch**. The Spot Fleet request type is `fleet`. When the request is fulfilled, requests of type `instance` are added, where the state is `active` and the status is `fulfilled`. ------ #### [ AWS CLI ] **To create a Spot Fleet request** Use the [request-spot-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/request-spot-fleet.html) command. ``` aws ec2 request-spot-fleet --spot-fleet-request-config file://config.json ``` For example configuration files, see [Example CLI configurations Spot Fleet](spot-fleet-examples.md). ------ #### [ PowerShell ] **To create a Spot Fleet request** Use the [Request-EC2SpotFleet](https://docs.aws.amazon.com/powershell/latest/reference/items/Request-EC2SpotFleet.html) cmdlet. The following example launches Spot Instances in a capacity-optimized fleet. ``` Request-EC2SpotFleet ` -SpotFleetRequestConfig_TargetCapacity 50 ` -SpotFleetRequestConfig_AllocationStrategy "CapacityOptimized" ` -SpotFleetRequestConfig_IamFleetRole "arn:aws:iam::123456789012:role/my-spot-fleet-role" ` -SpotFleetRequestConfig_LaunchTemplateConfig @($launchConfig) ``` Define the launch configuration as follows, setting the launch template and override properties that you need. For example configurations, see [Example CLI configurations Spot Fleet](spot-fleet-examples.md). ``` $lcSpec = Amazon.EC2.Model.FleetLaunchTemplateSpecification # To do - Set FleetLaunchTemplateSpecification properties $lcOverrides = New-Object Amazon.EC2.Model.LaunchTemplateOverrides # To do - Set LaunchTemplateOverrides properties $launchConfig = New-Object Amazon.EC2.Model.LaunchTemplateConfig $launchConfig.LaunchTemplateSpecification $lcSpec $launchConfig.Overrides @($lcOverrides) ``` ------ ## Create a Spot Fleet that replaces unhealthy Spot Instances Spot Fleet checks the health status of the Spot Instances in the fleet every two minutes. The health status of an instance is either `healthy` or `unhealthy`. Spot Fleet determines the health status of an instance by using the status checks provided by Amazon EC2. An instance is determined as `unhealthy` when the status of either the instance status check or the system status check is `impaired` for three consecutive health checks. For more information, see [Status checks for Amazon EC2 instances](monitoring-system-instance-status-check.md). You can configure your fleet to replace unhealthy Spot Instances. After enabling health check replacement, a Spot Instance is replaced when it is reported as `unhealthy`. The fleet could go below its target capacity for up to a few minutes while an unhealthy Spot Instance is being replaced. **Requirements** + Health check replacement is supported only for Spot Fleets that maintain a target capacity (fleets of type `maintain`), not for one-time Spot Fleets (fleets of type `request`). + Health check replacement is supported only for Spot Instances. This feature is not supported for On-Demand Instances. + You can configure your Spot Fleet to replace unhealthy instances only when you create it. + Users can use health check replacement only if they have permission to call the `ec2:DescribeInstanceStatus` action. ------ #### [ Console ] **To configure a Spot Fleet to replace unhealthy Spot Instances** 1. Follow the steps for creating a Spot Fleet in [Create a Spot Fleet request using defined parameters](#create-spot-fleet-advanced). 1. To configure the fleet to replace unhealthy Spot Instances, expand **Additional launch parameters**, and under **Health check**, select **Replace unhealthy instances**. To enable this option, you must first choose **Maintain target capacity**. ------ #### [ AWS CLI ] **To configure a Spot Fleet to replace unhealthy Spot Instances** Use the [request-spot-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/request-spot-fleet.html) command with the `ReplaceUnhealthyInstances` property of `SpotFleetRequestConfig`. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "lowestPrice", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "TargetCapacity": 10, "ReplaceUnhealthyInstances": true } } ``` ------ #### [ PowerShell ] **To configure a Spot Fleet request to replace unhealthy Spot Instances** Use the [Request-EC2SpotFleet](https://docs.aws.amazon.com/powershell/latest/reference/items/Request-EC2SpotFleet.html) cmdlet with the `-SpotFleetRequestConfig_ReplaceUnhealthyInstance` parameter. ``` -SpotFleetRequestConfig_ReplaceUnhealthyInstance $true ``` ------ # Tag a new or existing Spot Fleet request and the instances and volumes it launches To help categorize and manage your Spot Fleet requests and the instances and volumes that it launches, you can tag them with custom metadata. You can assign a tag to a Spot Fleet request when you create it, or afterward. Similarly, you can assign a tag to the instances and volumes when they're launched by the fleet, or afterward. When you tag a fleet request, the instances and volumes that are launched by the fleet are not automatically tagged. You need to explicitly tag the instances and volumes launched by the fleet. You can choose to assign tags to only the fleet request, or to only the instances launched by the fleet, or to only the volumes attached to the instances launched by the fleet, or to all of them. **Note** You can only tag volumes that are attached to On-Demand Instances. You can't tag volumes that are attached to Spot Instances. You can assign tags using the Amazon EC2 console or a command line tool. For more information about how tags work, see [Tag your Amazon EC2 resources](Using_Tags.md). **Topics** + [ ## Prerequisite ](#tag-spot-fleet-prereqs) + [ ## Tag a new Spot Fleet and the instances and volumes that it launches ](#tag-new-spot-fleet-and-resources) + [ ## Tag an existing Spot Fleet ](#tag-existing-spot-fleet) + [ ## View Spot Fleet request tags ](#view-spot-fleet-tags) ## Prerequisite Grant the user the permission to tag resources. For more information, see [Example: Tag resources](ExamplePolicies_EC2.md#iam-example-taggingresources). **To grant a user the permission to tag resources** Create an IAM policy that includes the following: + The `ec2:CreateTags` action. This grants the user permission to create tags. + The `ec2:RequestSpotFleet` action. This grants the user permission to create a Spot Fleet request. + For `Resource`, you must specify `"*"`. This allows users to tag all resource types. ------ #### [ JSON ] **** ``` { "Version":"2012-10-17", "Statement": [ { "Sid": "TagSpotFleetRequest", "Effect": "Allow", "Action": [ "ec2:CreateTags", "ec2:RequestSpotFleet" ], "Resource": "*" } ] } ``` ------ **Important** We currently do not support resource-level permissions for the `spot-fleet-request` resource. If you specify `spot-fleet-request` as a resource, you will get an unauthorized exception when you try to tag the fleet. The following example illustrates how *not* to set the policy. ``` { "Effect": "Allow", "Action": [ "ec2:CreateTags", "ec2:RequestSpotFleet" ], "Resource": "arn:aws:ec2:us-east-1:111122223333:spot-fleet-request/*" } ``` To provide access, add permissions to your users, groups, or roles: + Users and groups in AWS IAM Identity Center: Create a permission set. Follow the instructions in [Create a permission set](https://docs.aws.amazon.com//singlesignon/latest/userguide/howtocreatepermissionset.html) in the *AWS IAM Identity Center User Guide*. + Users managed in IAM through an identity provider: Create a role for identity federation. Follow the instructions in [Create a role for a third-party identity provider (federation)](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_roles_create_for-idp.html) in the *IAM User Guide*. + IAM users: + Create a role that your user can assume. Follow the instructions in [Create a role for an IAM user](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_roles_create_for-user.html) in the *IAM User Guide*. + (Not recommended) Attach a policy directly to a user or add a user to a user group. Follow the instructions in [Adding permissions to a user (console)](https://docs.aws.amazon.com//IAM/latest/UserGuide/id_users_change-permissions.html#users_change_permissions-add-console) in the *IAM User Guide*. ## Tag a new Spot Fleet and the instances and volumes that it launches **To tag a new Spot Fleet request and the instances and volumes that it launches using the console** 1. Follow the [Create a Spot Fleet request using defined parameters](create-spot-fleet.md#create-spot-fleet-advanced) procedure. 1. The way you add a tag depends on whether you manually configured the fleet or used a launch template. + If you manually configured the fleet, do the following: To add a tag, expand **Additional launch parameters**, choose **Create tag**, and enter the key and value for the tag. Repeat for each tag. For each tag, you can tag the Spot Fleet request and the instances with the same tag. To tag both, ensure that both **Instances** and **Fleet** are selected. To tag only the Spot Fleet request, clear **Instances**. To tag only the instances launched by the fleet, clear **Fleet**. **Note** When you manually configure a fleet, there is no option to tag volumes. Volume tags are only supported for volumes that are attached to On-Demand Instances. When you manually configure a fleet, you can't specify On-Demand Instances. + If you used a launch template, do the following: To add a tag to the fleet request, under **Tags**, choose **Create Tag**, and enter the key and value for the tag. Repeat for each tag. To tag the resources in your fleet, you must specify tags in the [launch template](create-launch-template.md). **To tag a new Spot Fleet request and the instances and volumes that it launches using the AWS CLI** To tag a Spot Fleet request when you create it, and to tag the instances and volumes when they are launched by the fleet, configure the Spot Fleet request configuration as follows: **Spot Fleet request tags:** + Specify the tags for the Spot Fleet request in `SpotFleetRequestConfig`. + For `ResourceType`, specify `spot-fleet-request`. If you specify another value, the fleet request will fail. + For `Tags`, specify the key-value pair. You can specify more than one key-value pair. **Instance tags:** + Specify the tags for the instances in `LaunchSpecifications`. + For `ResourceType`, specify `instance`. If you specify another value, the fleet request will fail. + For `Tags`, specify the key-value pair. You can specify more than one key-value pair. Alternatively, you can specify the tags for the instance in the [launch template](create-launch-template.md) that is referenced in the Spot Fleet request. **Volume tags:** + Specify the tags for the volumes in the [launch template](create-launch-template.md) that is referenced in the Spot Fleet request. Volume tagging in `LaunchSpecifications` is not supported. In the following example, the Spot Fleet request is tagged with two tags: Key=Environment and Value=Production, and Key=Cost-Center and Value=123. The instances that are launched by the fleet are tagged with one tag (which is the same as one of the tags for the Spot Fleet request): Key=Cost-Center and Value=123. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::111122223333:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-0123456789EXAMPLE", "InstanceType": "c4.large", "TagSpecifications": [ { "ResourceType": "instance", "Tags": [ { "Key": "Cost-Center", "Value": "123" } ] } ] } ], "SpotPrice": "5", "TargetCapacity": 2, "TerminateInstancesWithExpiration": true, "Type": "maintain", "ReplaceUnhealthyInstances": true, "InstanceInterruptionBehavior": "terminate", "InstancePoolsToUseCount": 1, "TagSpecifications": [ { "ResourceType": "spot-fleet-request", "Tags": [ { "Key": "Environment", "Value":"Production" }, { "Key": "Cost-Center", "Value":"123" } ] } ] } } ``` **To tag instances launched by a Spot Fleet using the AWS CLI** To tag instances when they are launched by the fleet, you can either specify the tags in the [launch template](create-launch-template.md) that is referenced in the Spot Fleet request, or you can specify the tags in the Spot Fleet request configuration as follows: + Specify the tags for the instances in `LaunchSpecifications`. + For `ResourceType`, specify `instance`. If you specify another value, the fleet request will fail. + For `Tags`, specify the key-value pair. You can specify more than one key-value pair. In the following example, the instances that are launched by the fleet are tagged with one tag: Key=Cost-Center and Value=123. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::111122223333:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-0123456789EXAMPLE", "InstanceType": "c4.large", "TagSpecifications": [ { "ResourceType": "instance", "Tags": [ { "Key": "Cost-Center", "Value": "123" } ] } ] } ], "SpotPrice": "5", "TargetCapacity": 2, "TerminateInstancesWithExpiration": true, "Type": "maintain", "ReplaceUnhealthyInstances": true, "InstanceInterruptionBehavior": "terminate", "InstancePoolsToUseCount": 1 } } ``` **To tag volumes attached to On-Demand Instances launched by a Spot Fleet using the AWS CLI** To tag volumes when they are created by the fleet, you must specify the tags in the [launch template](create-launch-template.md) that is referenced in the Spot Fleet request. **Note** Volume tags are only supported for volumes that are attached to On-Demand Instances. You can't tag volumes that are attached to Spot Instances. Volume tagging in `LaunchSpecifications` is not supported. ## Tag an existing Spot Fleet **To tag an existing Spot Fleet request using the console** After you create a Spot Fleet request, you can add tags to the fleet request using the console. 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose the **Tags** tab and choose **Create Tag**. **To tag an existing Spot Fleet request using the AWS CLI** You can use the [create-tags](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-tags.html) command to tag existing resources. In the following example, the existing Spot Fleet request is tagged with Key=purpose and Value=test. ``` aws ec2 create-tags \ --resources sfr-11112222-3333-4444-5555-66666EXAMPLE \ --tags Key=purpose,Value=test ``` ## View Spot Fleet request tags **To view Spot Fleet request tags using the console** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request and choose the **Tags** tab. **To describe Spot Fleet request tags** Use the [describe-tags](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-tags.html) command to view the tags for the specified resource. In the following example, you describe the tags for the specified Spot Fleet request. ``` aws ec2 describe-tags \ --filters "Name=resource-id,Values=sfr-11112222-3333-4444-5555-66666EXAMPLE" ``` ``` { "Tags": [ { "Key": "Environment", "ResourceId": "sfr-11112222-3333-4444-5555-66666EXAMPLE", "ResourceType": "spot-fleet-request", "Value": "Production" }, { "Key": "Another key", "ResourceId": "sfr-11112222-3333-4444-5555-66666EXAMPLE", "ResourceType": "spot-fleet-request", "Value": "Another value" } ] } ``` You can also view the tags of a Spot Fleet request by describing the Spot Fleet request. Use the [describe-spot-fleet-requests](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-spot-fleet-requests.html) command to view the configuration of the specified Spot Fleet request, which includes any tags that were specified for the fleet request. ``` aws ec2 describe-spot-fleet-requests \ --spot-fleet-request-ids sfr-11112222-3333-4444-5555-66666EXAMPLE ``` ``` { "SpotFleetRequestConfigs": [ { "ActivityStatus": "fulfilled", "CreateTime": "2020-02-13T02:49:19.709Z", "SpotFleetRequestConfig": { "AllocationStrategy": "capacityOptimized", "OnDemandAllocationStrategy": "lowestPrice", "ExcessCapacityTerminationPolicy": "Default", "FulfilledCapacity": 2.0, "OnDemandFulfilledCapacity": 0.0, "IamFleetRole": "arn:aws:iam::111122223333:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-0123456789EXAMPLE", "InstanceType": "c4.large" } ], "TargetCapacity": 2, "OnDemandTargetCapacity": 0, "Type": "maintain", "ReplaceUnhealthyInstances": false, "InstanceInterruptionBehavior": "terminate" }, "SpotFleetRequestId": "sfr-11112222-3333-4444-5555-66666EXAMPLE", "SpotFleetRequestState": "active", "Tags": [ { "Key": "Environment", "Value": "Production" }, { "Key": "Another key", "Value": "Another value" } ] } ] } ``` # Describe a Spot Fleet request, its instances, and event history You can describe your Spot Fleet configuration, the instances in your Spot Fleet, and the event history of your Spot Fleet. ------ #### [ Console ] **To describe your Spot Fleet** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. The ID begins with **sfr-**. To see the configuration details, choose **Description**. 1. To list the Spot Instances for the Spot Fleet, choose **Instances**. 1. To view the history for the Spot Fleet, choose **History**. ------ #### [ AWS CLI ] **To describe your Spot Fleet request** Use the [describe-spot-fleet-requests](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-spot-fleet-requests.html) command. ``` aws ec2 describe-spot-fleet-requests \ --spot-fleet-request-ids sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE ``` **To describe the running instances for the specified Spot Fleet request** Use the [describe-spot-fleet-instances](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-spot-fleet-instances.html) command. ``` aws ec2 describe-spot-fleet-instances \ --spot-fleet-request-id sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE ``` **To describe the event history for the specified Spot Fleet request** Use the [describe-spot-fleet-request-history](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-spot-fleet-request-history.html) command. ``` aws ec2 describe-spot-fleet-request-history \ --spot-fleet-request-id sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --start-time 2024-05-18T00:00:00Z ``` ------ #### [ PowerShell ] **To describe your Spot Fleet request** Use the [Get-EC2SpotFleetRequest](https://docs.aws.amazon.com/powershell/latest/reference/items/Get-EC2SpotFleetRequest.html) cmdlet. ``` Get-EC2SpotFleetRequest ``` **To describe the running instances for the specified Spot Fleet request** Use the [Get-EC2SpotFleetInstance](https://docs.aws.amazon.com/powershell/latest/reference/items/Get-EC2SpotFleetInstance.html) cmdlet. ``` Get-EC2SpotFleetInstance ` -SpotFleetRequestId "sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ``` **To describe the event history for the specified Spot Fleet request** Use the [Get-EC2SpotFleetRequestHistory](https://docs.aws.amazon.com/powershell/latest/reference/items/Get-EC2SpotFleetRequestHistory.html) cmdlet. ``` Get-EC2SpotFleetRequestHistory ` -SpotFleetRequestId "sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -UtcStartTime 2024-05-18T00:00:00Z ``` ------ # Modify a Spot Fleet request You can modify an active Spot Fleet request to complete the following tasks: + Increase the total target capacity and On-Demand portion + Decrease the total target capacity and On-Demand portion When you increase the total target capacity, the Spot Fleet launches additional Spot Instances according to the [allocation strategy](ec2-fleet-allocation-strategy.md) for its Spot Fleet request. When you increase the On-Demand portion, the Spot Fleet launches additional On-Demand Instances. When you decrease the total target capacity, the Spot Fleet cancels any open requests that exceed the new target capacity. You can request that the Spot Fleet terminate Spot Instances until the size of the fleet reaches the new target capacity. If the allocation strategy is `diversified`, the Spot Fleet terminates instances across the pools. Alternatively, you can request that the Spot Fleet keep the fleet at its current size, but not replace any Spot Instances that are interrupted or that you terminate manually. **Considerations** + You can't modify a one-time Spot Fleet request. You can only modify a Spot Fleet request if you selected **Maintain target capacity** when you created the Spot Fleet request. + When a Spot Fleet terminates an instance because the target capacity was decreased, the instance receives a Spot Instance interruption notice. ------ #### [ Console ] **To modify a Spot Fleet request** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose **Actions**, **Modify target capacity**. 1. In **Modify target capacity**, do the following: 1. Enter the new target capacity and On-Demand portion. 1. (Optional) If you are decreasing the target capacity but want to keep the fleet at its current size, clear **Terminate instances**. 1. Choose **Submit**. ------ #### [ AWS CLI ] **To modify a Spot Fleet request** Use the [modify-spot-fleet-request](https://docs.aws.amazon.com/cli/latest/reference/ec2/modify-spot-fleet-request.html) command to update the target capacity of the specified Spot Fleet request. ``` aws ec2 modify-spot-fleet-request \ --spot-fleet-request-id sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --target-capacity 20 ``` You can modify the previous command as follows to decrease the target capacity of the specified Spot Fleet without terminating any Spot Instances as a result. ``` aws ec2 modify-spot-fleet-request \ --spot-fleet-request-id sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --target-capacity 10 \ --excess-capacity-termination-policy NoTermination ``` ------ #### [ PowerShell ] **To modify a Spot Fleet request** Use the [Edit-EC2SpotFleetRequest](https://docs.aws.amazon.com/powershell/latest/reference/items/Edit-EC2SpotFleetRequest.html) cmdlet to update the target capacity of the specified Spot Fleet request. ``` Edit-EC2SpotFleetRequest ` -SpotFleetRequestId "sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -TargetCapacity 20 ``` You can modify the previous command as follows to decrease the target capacity of the specified Spot Fleet without terminating any Spot Instances as a result. ``` Edit-EC2SpotFleetRequest ` -SpotFleetRequestId "sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -TargetCapacity 20 ` -ExcessCapacityTerminationPolicy "NoTermination" ``` ------ # Cancel (delete) a Spot Fleet request If you no longer require a Spot Fleet, you can cancel the Spot Fleet request, which deletes the request. After you cancel a fleet request, all Spot requests associated with the fleet are also canceled, so that no new Spot Instances are launched. When you cancel a Spot Fleet request, you must also specify if you want to terminate all of its instances. These include both On-Demand Instances and Spot Instances. **Warning** **Terminating an instance is permanent and irreversible.** After you terminate an instance, you can no longer connect to it, and it can't be recovered. All attached Amazon EBS volumes that are configured to be deleted on termination are also permanently deleted and can't be recovered. All data stored on instance store volumes is permanently lost. For more information, see [How instance termination works](how-ec2-instance-termination-works.md). Before you terminate an instance, ensure that you have backed up all data that you need to retain after the termination to persistent storage. If you specify that the instances must be terminated when the fleet request is canceled, the fleet request enters the `cancelled_terminating` state. Otherwise, it enters the `cancelled_running` state, and the instances continue to run until they are interrupted or you terminate them manually. **Restrictions** + You can cancel up to 100 fleets in a single request. If you exceed the specified number, no fleets are canceled. ------ #### [ Console ] **To cancel (delete) a Spot Fleet request** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose **Actions**, **Cancel request**. 1. In the **Cancel Spot request** dialog box, do the following: 1. To terminate the associated instances at the same time as canceling the Spot Fleet request, leave the **Terminate instances** checkbox selected. To cancel the Spot Fleet request without terminating the associated instances, clear the **Terminate instances** checkbox. 1. Choose **Confirm**. ------ #### [ AWS CLI ] **To cancel (delete) a Spot Fleet request and terminate its instances** Use the [cancel-spot-fleet-requests](https://docs.aws.amazon.com/cli/latest/reference/ec2/cancel-spot-fleet-requests.html) command with the `--terminate-instances` option. ``` aws ec2 cancel-spot-fleet-requests \ --spot-fleet-request-ids sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --terminate-instances ``` The following is example output. ``` { "SuccessfulFleetRequests": [ { "SpotFleetRequestId": "sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE", "CurrentSpotFleetRequestState": "cancelled_terminating", "PreviousSpotFleetRequestState": "active" } ], "UnsuccessfulFleetRequests": [] } ``` **To cancel (delete) a Spot Fleet request without terminating its instances** Modify the previous example by using the `--no-terminate-instances` option instead. ``` aws ec2 cancel-spot-fleet-requests \ --spot-fleet-request-ids sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE \ --no-terminate-instances ``` The following is example output. ``` { "SuccessfulFleetRequests": [ { "SpotFleetRequestId": "sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE", "CurrentSpotFleetRequestState": "cancelled_running", "PreviousSpotFleetRequestState": "active" } ], "UnsuccessfulFleetRequests": [] } ``` ------ #### [ PowerShell ] **To cancel (delete) a Spot Fleet request and terminate its instances** Use the [Stop-EC2SpotFleetRequest](https://docs.aws.amazon.com/powershell/latest/reference/items/Stop-EC2SpotFleetRequest.html) cmdlet with the `-TerminateInstance` parameter. ``` Stop-EC2SpotFleetRequest ` -SpotFleetRequestId "sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -TerminateInstance $true ``` **To cancel (delete) a Spot Fleet request without terminating its instances** Modify the previous example by changing the value of the `-TerminateInstance` parameter. ``` Stop-EC2SpotFleetRequest ` -SpotFleetRequestId "sfr-73fbd2ce-aa30-494c-8788-1cee4EXAMPLE" ` -TerminateInstance $false ``` ------ # Understand automatic scaling for Spot Fleet *Automatic scaling* enables your Spot Fleet to increase or decrease its target capacity based on demand. With automatic scaling, a Spot Fleet can either launch instances (scale out) or terminate instances (scale in) within a specified range, in response to one or more scaling policies. Automatic scaling for Spot Fleet is made possible by a combination of the Amazon EC2, Amazon CloudWatch, and Application Auto Scaling APIs. Spot Fleet requests are created with Amazon EC2, alarms are created with CloudWatch, and scaling policies are created with Application Auto Scaling. **Types of automatic scaling** Spot Fleet supports the following types of automatic scaling: + [Target tracking scaling](spot-fleet-target-tracking.md) – Increase or decrease
 the current capacity of the fleet by targeting a value for a specific metric. This is similar to the way that your thermostat maintains the temperature of your home—you select the desired temperature and the thermostat does the rest. + [Step scaling](spot-fleet-step-scaling.md) – Increase or decrease the current capacity of the fleet based on a set of scaling adjustments, known as step adjustments, that vary based on the size of the alarm breach. + [Scheduled scaling](spot-fleet-scheduled-scaling.md) – Increase or decrease the current capacity of the fleet based on the date and time. ## Considerations When using automatic scaling for your Spot Fleet, consider the following: + **Instance weighting** – If you're using [instance weighting](ec2-fleet-instance-weighting.md), keep in mind that Spot Fleet can exceed the target capacity as needed. Fulfilled capacity can be a floating-point number but target capacity must be an integer, so Spot Fleet rounds up to the next integer. You must take these behaviors into account when you look at the outcome of a scaling policy when an alarm is triggered. For example, suppose that the target capacity is 30, the fulfilled capacity is 30.1, and the scaling policy subtracts 1. When the alarm is triggered, the automatic scaling process subtracts 1 from 30.1 to get 29.1 and then rounds it up to 30, so no scaling action is taken. As another example, suppose that you selected instance weights of 2, 4, and 8, and a target capacity of 10, but no weight 2 instances were available so Spot Fleet provisioned instances of weights 4 and 8 for a fulfilled capacity of 12. If the scaling policy decreases target capacity by 20% and an alarm is triggered, the automatic scaling process subtracts 12\$10.2 from 12 to get 9.6 and then rounds it up to 10, so no scaling action is taken. + **Cooldown period** – The scaling policies that you create for Spot Fleet support a cooldown period. This is the number of seconds after a scaling activity completes where previous trigger-related scaling activities can influence future scaling events. For scale-out policies, while the cooldown period is in effect, the capacity that has been added by the previous scale-out event that initiated the cooldown is calculated as part of the desired capacity for the next scale out. The intention is to continuously (but not excessively) scale out. For scale in policies, the cooldown period is used to block subsequent scale in requests until it has expired. The intention is to scale in conservatively to protect your application's availability. However, if another alarm triggers a scale-out policy during the cooldown period after a scale-in, automatic scaling scales out your scalable target immediately. + **Use detailed monitoring** – We recommend that you scale based on instance metrics with a 1-minute frequency because that ensures a faster response to utilization changes. Scaling on metrics with a 5-minute frequency can result in slower response time and scaling on stale metric data. To send metric data for your instances to CloudWatch in 1-minute periods, you must specifically enable detailed monitoring. For more information, see [Manage detailed monitoring for your EC2 instances](manage-detailed-monitoring.md) and [Create a Spot Fleet request using defined parameters](create-spot-fleet.md#create-spot-fleet-advanced). + **AWS CLI** – If you use the AWS CLI for configuring scaling for Spot Fleet, you'll use the [application-autoscaling](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling) commands. # IAM permissions required for Spot Fleet automatic scaling Automatic scaling for Spot Fleet is made possible by a combination of the Amazon EC2, Amazon CloudWatch, and Application Auto Scaling APIs. Spot Fleet requests are created with Amazon EC2, alarms are created with CloudWatch, and scaling policies are created with Application Auto Scaling. In addition to the [IAM permissions required for using Spot Fleet](spot-fleet-prerequisites.md#spot-fleet-iam-users) and Amazon EC2, the user that accesses the fleet scaling settings must have the appropriate permissions for the services that support automatic scaling. To use automatic scaling for Spot Fleet, users must have permissions to use the actions shown in the following example policy. ------ #### [ JSON ] **** ``` { "Version":"2012-10-17", "Statement": [ { "Effect": "Allow", "Action": [ "application-autoscaling:*", "ec2:DescribeSpotFleetRequests", "ec2:ModifySpotFleetRequest", "cloudwatch:DeleteAlarms", "cloudwatch:DescribeAlarmHistory", "cloudwatch:DescribeAlarms", "cloudwatch:DescribeAlarmsForMetric", "cloudwatch:GetMetricStatistics", "cloudwatch:ListMetrics", "cloudwatch:PutMetricAlarm", "cloudwatch:DisableAlarmActions", "cloudwatch:EnableAlarmActions", "iam:CreateServiceLinkedRole", "sns:CreateTopic", "sns:Subscribe", "sns:Get*", "sns:List*" ], "Resource": "*" } ] } ``` ------ You can also create your own IAM policies that allow more fine-grained permissions for calls to the Application Auto Scaling API. For more information, see [Identity and Access Management for Application Auto Scaling](https://docs.aws.amazon.com/autoscaling/application/userguide/auth-and-access-control.html) in the *Application Auto Scaling User Guide*. The Application Auto Scaling service also needs permission to describe your Spot Fleet and CloudWatch alarms, and permissions to modify your Spot Fleet target capacity on your behalf. If you enable automatic scaling for your Spot Fleet, it creates a service-linked role named `AWSServiceRoleForApplicationAutoScaling_EC2SpotFleetRequest`. This service-linked role grants Application Auto Scaling permission to describe the alarms for your policies, to monitor the current capacity of the fleet, and to modify the capacity of the fleet. The original managed Spot Fleet role for Application Auto Scaling was `aws-ec2-spot-fleet-autoscale-role`, but it is no longer required. The service-linked role is the default role for Application Auto Scaling. For more information, see [Service-linked roles for Application Auto Scaling](https://docs.aws.amazon.com/autoscaling/application/userguide/application-auto-scaling-service-linked-roles.html) in the *Application Auto Scaling User Guide*. # Target tracking scaling: Scale Spot Fleet by targeting a value for a specific metric With target tracking scaling, you create a target tracking scaling policy by selecting a metric and setting a target value. Spot Fleet then creates and manages the CloudWatch alarms that trigger the scaling policy, and calculates the scaling adjustment based on the chosen metric and target value. The scaling policy adjusts capacity by adding or removing instances as needed to keep the metric at, or close to, the specified target value. A target tracking policy not only keeps the metric close to the target value, but also adjusts to the fluctuations in the metric due to a fluctuating load pattern and minimizes rapid capacity fluctuations. You can create multiple target tracking scaling policies for a Spot Fleet, provided each policy uses a different metric. The fleet scales based on the policy that specifies the largest fleet capacity. This allows you to cover multiple scenarios to ensure sufficient capacity for your application workloads. To ensure application availability, the fleet scales out proportionally to the metric as fast as it can, but scales in more gradually. When a Spot Fleet terminates a Spot Instance because the target capacity was decreased, the instance receives a Spot Instance interruption notice. **Note** Do not edit or delete the CloudWatch alarms that Spot Fleet manages for a target tracking scaling policy. Spot Fleet deletes the alarms automatically when you delete the target tracking scaling policy. **Prerequisites** + The Spot Fleet request must have a request type of `maintain`. Automatic scaling is not supported for requests of type `request`. + Configure the [IAM permissions required for Spot Fleet automatic scaling](spot-fleet-auto-scaling-IAM.md). + Review the [Considerations](spot-fleet-automatic-scaling.md#considerations-for-spot-fleet-automatic-scaling). **To configure a target tracking policy** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose the **Auto Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Auto Scaling** section. 1. If automatic scaling is not configured, choose **Configure**. 1. Use **Scale capacity between** to set the minimum and maximum capacity for your fleet. Automatic scaling does not scale your fleet below the minimum capacity or above the maximum capacity. 1. For **Policy name**, enter a name for the policy. 1. Choose a **Target metric**. 1. Enter a **Target value** for the metric. 1. For **Cooldown period**, specify a new value (in seconds) or keep the default. 1. (Optional) To omit creating a scale-in policy based on the current configuration, select **Disable scale-in**. You can create a scale-in policy using a different configuration. 1. Choose **Save**. **To configure a target tracking policy using the AWS CLI** 1. Register the Spot Fleet request as a scalable target using the [register-scalable-target](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/register-scalable-target.html) command. 1. Create a scaling policy using the [put-scaling-policy](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/put-scaling-policy.html) command. # Step scaling: Scale Spot Fleet using step scaling policies With step scaling policies, you specify CloudWatch alarms to trigger the scaling process. For example, if you want to scale out when CPU utilization reaches a certain level, create an alarm using the `CPUUtilization` metric provided by Amazon EC2. When you create a step scaling policy, you must specify one of the following scaling adjustment types: + **Add** – Increase the target capacity of the fleet by a specified number of capacity units or a specified percentage of the current capacity. + **Remove** – Decrease the target capacity of the fleet by a specified number of capacity units or a specified percentage of the current capacity. + **Set to** – Set the target capacity of the fleet to the specified number of capacity units. When an alarm is triggered, the automatic scaling process calculates the new target capacity using the fulfilled capacity and the scaling policy, and then updates the target capacity accordingly. For example, suppose that the target capacity and fulfilled capacity are 10 and the scaling policy adds 1. When the alarm is triggered, the automatic scaling process adds 1 to 10 to get 11, so Spot Fleet launches 1 instance. When a Spot Fleet terminates a Spot Instance because the target capacity was decreased, the instance receives a Spot Instance interruption notice. **Prerequisites** + The Spot Fleet request must have a request type of `maintain`. Automatic scaling is not supported for requests of type `request`. + Configure the [IAM permissions required for Spot Fleet automatic scaling](spot-fleet-auto-scaling-IAM.md). + Consider which CloudWatch metrics are important to your application. You can create CloudWatch alarms based on metrics provided by AWS or your own custom metrics. + For the AWS metrics that you will use in your scaling policies, enable CloudWatch metrics collection if the service that provides the metrics does not enable it by default. + Review the [Considerations](spot-fleet-automatic-scaling.md#considerations-for-spot-fleet-automatic-scaling). **To create a CloudWatch alarm** 1. Open the CloudWatch console at [https://console.aws.amazon.com/cloudwatch/](https://console.aws.amazon.com/cloudwatch/). 1. In the navigation pane, expand **Alarms** and choose **All alarms**. 1. Choose **Create alarm**. 1. On the **Specify metric and conditions** page, choose **Select metric**. 1. Choose **EC2 Spot**, then **Fleet Request Metrics**, and then select a metric (for example, **TargetCapacity**), and then choose **Select metric**. The **Specify metric and conditions** page appears, showing a graph and other information about the metric you selected. 1. For **Period**, choose the evaluation period for the alarm, for example, **1 minute**. When evaluating the alarm, each period is aggregated into one data point. **Note** A shorter period creates a more sensitive alarm. 1. For **Conditions**, define the alarm by defining the threshold condition. For example, you can define a threshold to trigger the alarm whenever the value of the metric is greater than or equal to 80 percent. 1. Under **Additional configuration**, for **Datapoints to alarm**, specify how many datapoints (evaluation periods) must be in the ALARM state to trigger the alarm, for example, 1 evaluation period or 2 out of 3 evaluation periods. This creates an alarm that goes to ALARM state if that many consecutive periods are breaching. For more information, see [Evaluating an alarm](https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/AlarmThatSendsEmail.html#alarm-evaluation) in the *Amazon CloudWatch User Guide*. 1. For **Missing data treatment**, choose one of the options (or leave the default of **Treat missing data as missing**). For more information, see [Configuring how CloudWatch alarms treat missing data](https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/AlarmThatSendsEmail.html#alarms-and-missing-data) in the *Amazon CloudWatch User Guide*. 1. Choose **Next**. 1. (Optional) To receive notification of a scaling event, for **Notification**, you can choose or create the Amazon SNS topic you want to use to receive notifications. Otherwise, you can delete the notification now and add one later as needed. 1. Choose **Next**. 1. Under **Add name and description**, enter a name and description for the alarm and choose **Next**. 1. Choose **Create alarm**. **To configure a step scaling policy for your Spot Fleet** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose the **Auto Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Auto Scaling** section. 1. If automatic scaling is not configured, choose **Configure**. 1. Use **Scale capacity between** to set the minimum and maximum capacity for your fleet. Scaling policies do not scale your fleet below the minimum capacity or above the maximum capacity. 1. Under **Scaling policies**, for **Policy type**, choose **Step scaling policy**. 1. Initially, **Scaling policies** contain step scaling policies named **ScaleUp** and **ScaleDown**. You can complete these policies, or choose **Remove policy** to delete them. You can also choose **Add policy**. 1. To define a policy, do the following: 1. For **Policy name**, enter a name for the policy. 1. For **Policy trigger**, select an existing alarm, or choose **Create alarm** to open the Amazon CloudWatch console and create an alarm. 1. For **Modify capacity**, define the amount by which to scale and the lower and upper bound of the step adjustment. You can add or remove a specific number of instances or a percentage of the existing fleet size, or set the fleet to an exact size. For example, to create a step scaling policy that increases the capacity of the fleet by 30 percent, choose **Add**, enter **30** in the next field, and then choose **percent**. By default, the lower bound for an add policy is the alarm threshold and the upper bound is positive (\$1) infinity. By default, the upper bound for a remove policy is the alarm threshold and the lower bound is negative (-) infinity. 1. (Optional) To add another step, choose **Add step**. 1. For **Cooldown period**, specify a new value (in seconds) or keep the default. 1. Choose **Save**. **To configure step scaling policies for your Spot Fleet using the AWS CLI** 1. Register the Spot Fleet request as a scalable target using the [register-scalable-target](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/register-scalable-target.html) command. 1. Create a scaling policy using the [put-scaling-policy](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/put-scaling-policy.html) command. 1. Create an alarm that triggers the scaling policy using the [put-metric-alarm](https://docs.aws.amazon.com/cli/latest/reference/cloudwatch/put-metric-alarm.html) command. # Scheduled scaling: Scale Spot Fleet on a schedule Scaling your fleet on a schedule enables you to scale your application in response to predictable changes in demand. By creating *scheduled actions*, you can instruct Spot Fleet to perform scaling activities at specific times. To create a scheduled action, you must specify an existing Spot Fleet, the time when the scaling activity must occur, and the desired minimum and maximum capacity. Scheduled actions can be configured to scale once or on a recurring schedule. If you needs change, you can edit or delete scheduled actions. **Prerequisites** + Scheduled actions can only be created for existing Spot Fleets. You can't create a scheduled action when you create a Spot Fleet. + The Spot Fleet request must have a request type of `maintain`. Automatic scaling is not supported for requests of type `request`. + Configure the [IAM permissions required for Spot Fleet automatic scaling](spot-fleet-auto-scaling-IAM.md). + Review the [Considerations](spot-fleet-automatic-scaling.md#considerations-for-spot-fleet-automatic-scaling). **To create a one-time scheduled action** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose the **Scheduled Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Scheduled Scaling** section. 1. Choose **Create scheduled action**. 1. For **Name**, specify a name for the scheduled action. 1. Enter a value for **Minimum capacity**, **Maximum capacity**, or both. 1. For **Recurrence**, choose **Once**. 1. (Optional) Choose a date and time for **Start time**, **End time**, or both. 1. Choose **Create**. **To create a recurring scheduled action** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose the **Scheduled Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Scheduled Scaling** section. 1. For **Name**, specify a name for the scheduled action. 1. Enter a value for **Minimum capacity**, **Maximum capacity**, or both. 1. For **Recurrence**, choose one of the predefined schedules (for example, **Every day**), or choose **Custom** and enter a cron expression. For more information about the cron expressions supported by scheduled scaling, see [Cron expressions](https://docs.aws.amazon.com/eventbridge/latest/userguide/eb-scheduled-rule-pattern.html#eb-cron-expressions) in the *Amazon EventBridge User Guide*. 1. (Optional) Choose a date and time for **Start time**, **End time**, or both. 1. Choose **Submit**. **To edit a scheduled action** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose the **Scheduled Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Scheduled Scaling** section. 1. Select the scheduled action and choose **Actions**, **Edit**. 1. Make the needed changes and choose **Submit**. **To delete a scheduled action** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. In the navigation pane, choose **Spot Requests**. 1. Select your Spot Fleet request. 1. Choose the **Scheduled Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Scheduled Scaling** section. 1. Select the scheduled action and choose **Actions**, **Delete**. 1. When prompted for confirmation, choose **Delete**. **To manage scheduled scaling using the AWS CLI** Use the following commands: + [put-scheduled-action](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/put-scheduled-action.html) + [describe-scheduled-actions](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/describe-scheduled-actions.html) + [delete-scheduled-action](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/delete-scheduled-action.html) # Monitor your EC2 Fleet or Spot Fleet Effective monitoring of your EC2 Fleet or Spot Fleet is essential for maintaining optimal performance and ensuring reliability. There are various tools to help you achieve this, including Amazon CloudWatch and Amazon EventBridge, which are covered in this topic. With CloudWatch, you can collect and track metrics, set alarms, and automatically react to changes in your fleet’s status. With EventBridge, you can monitor and respond programmatically to events emitted by your fleet. By defining rules in EventBridge, you can automate responses to specific fleet events, such as instance termination or fleet state changes, improving your operational efficiency. **Topics** + [ # Monitor your EC2 Fleet or Spot Fleet using CloudWatch ](ec2-fleet-cloudwatch-metrics.md) + [ # Monitor and programmatically respond to the events emitted by your EC2 Fleet or Spot Fleet using Amazon EventBridge ](monitor-ec2-fleet-using-eventbridge.md) # Monitor your EC2 Fleet or Spot Fleet using CloudWatch You can monitor your EC2 Fleet or Spot Fleet using the Amazon CloudWatch metrics described in this section. **Important** To ensure accuracy, we recommend that you enable detailed monitoring when using these metrics. For more information, see [Manage detailed monitoring for your EC2 instances](manage-detailed-monitoring.md). For more information about using CloudWatch, see [Monitor your instances using CloudWatch](using-cloudwatch.md). ## EC2 Fleet and Spot Fleet metrics The `AWS/EC2Spot` namespace includes the following metrics for your fleet, plus the CloudWatch metrics for the Spot Instances in your fleet. For more information, see [Instance metrics](viewing_metrics_with_cloudwatch.md#ec2-cloudwatch-metrics). | Metric | Description | | --- | --- | | AvailableInstancePoolsCount | The Spot capacity pools specified in the fleet request. Units: Count | | BidsSubmittedForCapacity | The capacity for which Amazon EC2 has submitted fleet requests. Units: Count | | EligibleInstancePoolCount | The Spot capacity pools specified in the fleet request where Amazon EC2 can fulfill requests. Amazon EC2 does not fulfill requests in pools where the maximum price you're willing to pay for Spot Instances is less than the Spot price or the Spot price is greater than the price for On-Demand Instances. Units: Count | | FulfilledCapacity | The capacity that Amazon EC2 has fulfilled. Units: Count | | MaxPercentCapacityAllocation | The maximum value of `PercentCapacityAllocation` across all fleet pools specified in the fleet request. Units: Percent | | PendingCapacity | The difference between `TargetCapacity` and `FulfilledCapacity`. Units: Count | | PercentCapacityAllocation | The capacity allocated for the Spot capacity pool for the specified dimensions. To get the maximum value recorded across all Spot capacity pools, use `MaxPercentCapacityAllocation`. Units: Percent | | TargetCapacity | The target capacity of the fleet request. Units: Count | | TerminatingCapacity | The capacity that is being terminated because the provisioned capacity is greater than the target capacity. Units: Count | If the unit of measure for a metric is `Count`, the most useful statistic is `Average`. ## EC2 Fleet and Spot Fleet dimensions To filter the data for your fleet, use the following dimensions. | Dimensions | Description | | --- | --- | | AvailabilityZone | Filter the data by Availability Zone. | | FleetRequestId | Filter the data by fleet request. | | InstanceType | Filter the data by instance type. | ## View the CloudWatch metrics for your EC2 Fleet or Spot Fleet You can view the CloudWatch metrics for your fleet using the Amazon CloudWatch console. These metrics are displayed as monitoring graphs. These graphs show data points if the fleet is active. Metrics are grouped first by namespace, and then by the various combinations of dimensions within each namespace. For example, you can view all fleet metrics or fleet metrics groups by fleet request ID, instance type, or Availability Zone. **To view fleet metrics** 1. Open the CloudWatch console at [https://console.aws.amazon.com/cloudwatch/](https://console.aws.amazon.com/cloudwatch/). 1. In the navigation pane, expand **Metrics**, and choose **All metrics**. 1. Choose the **EC2 Spot** namespace. **Note** If the **EC2 Spot** namespace is not displayed, there are two reasons for this. Either you have never used EC2 Fleet or Spot Fleet in the Region—only the AWS services that you're using send metrics to Amazon CloudWatch. Or, if you have used EC2 Fleet or Spot Fleet in the Region, but not for the past two weeks, the namespace does not appear. 1. To filter the metrics by dimension, choose one of the following: + **Fleet Request Metrics** – Group by fleet request + **By Availability Zone** – Group by fleet request and Availability Zone + **By Instance Type** – Group by fleet request and instance type + **By Availability Zone/Instance Type** – Group by fleet request, Availability Zone, and instance type 1. To view the data for a metric, select the checkbox next to the metric. # Monitor and programmatically respond to the events emitted by your EC2 Fleet or Spot Fleet using Amazon EventBridge When the state of an EC2 Fleet or Spot Fleet changes, it emits a notification. The notification is made available as an event that is sent to Amazon EventBridge (formerly known as Amazon CloudWatch Events). Events are emitted on a best effort basis. You can use Amazon EventBridge to create rules that trigger programmatic actions in response to an event. For example, you can create two EventBridge rules: one triggered when a fleet state changes, and another triggered when an instance in the fleet is terminated. In this example, you can configure the first rule so that, if the fleet state changes, the rule invokes an SNS topic, sending an email notification to you. You can configure the second rule so that, if an instance in the fleet is terminated, the rule invokes a Lambda function to launch a new instance. **Note** Only fleets of type `maintain` and `request` emit events. Fleets of type `instant` do not emit events because they submit synchronous one-time requests, and the state of the fleet is known immediately in the response. To use Amazon EventBridge to monitor fleet events, the request type must be `maintain` or `request`. For instructions on how to describe a fleet's event history, see [Describe the event history for your EC2 Fleet](describe-ec2-fleet.md#describe-ec2-fleet-event-history). **Topics** + [ ## Create Amazon EventBridge rules to monitor EC2 Fleet or Spot Fleet events ](#fleet_create-eventbridge-rules) + [ ## EC2 Fleet event types ](#ec2-fleet-event-types) + [ ## Spot Fleet event types ](#spot-fleet-event-types) ## Create Amazon EventBridge rules to monitor EC2 Fleet or Spot Fleet events When a state change notification is emitted for an EC2 Fleet or Spot Fleet, it is sent as an event to Amazon EventBridge as a JSON file. If EventBridge detects an event pattern that matches a pattern defined in a rule, EventBridge invokes the target (or targets) specified in the rule. You can write EventBridge rules to automate actions based on matching event patterns. The following fields in the event form the event pattern that is defined in the rule: `"source": "aws.ec2fleet"` Identifies that the event is from EC2 Fleet. `"detail-type": "EC2 Fleet State Change"` Identifies the event type. `"detail": { "sub-type": "submitted" }` Identifies the event sub-type. For the list of EC2 Fleet and Spot Fleet events and example event data, see [EC2 Fleet event types](#ec2-fleet-event-types) and [Spot Fleet event types](#spot-fleet-event-types). **Topics** + [ ### Create an EventBridge rule to send a notification ](#eventbridge-send-notification) + [ ### Create an EventBridge rule to trigger a Lambda function ](#eventbridge-trigger-lambda) ### Create an EventBridge rule to send a notification The following example creates an EventBridge rule to send an email, text message, or mobile push notification every time that Amazon EC2 emits an EC2 Fleet state change notification. The signal in this example is emitted as an `EC2 Fleet State Change` event, which triggers the action defined by the rule. **Prerequisite** Before creating the EventBridge rule, you must create the Amazon SNS topic for the email, text message, or mobile push notification. **To create an EventBridge rule to send a notification when an EC2 Fleet state changes** 1. Open the Amazon EventBridge console at [https://console.aws.amazon.com/events/](https://console.aws.amazon.com/events/). 1. Choose **Create rule**. 1. For **Define rule detail**, do the following: 1. Enter a **Name** for the rule, and, optionally, a description. A rule can't have the same name as another rule in the same Region and on the same event bus. 1. For **Event bus**, choose **default**. When an AWS service in your account generates an event, it always goes to your account's default event bus. 1. For **Rule type**, choose **Rule with an event pattern**. 1. Choose **Next**. 1. For **Build event pattern**, do the following: 1. For **Event source**, choose **AWS events or EventBridge partner events**. 1. For **Event pattern**, for this example you’ll specify the following event pattern to match the `EC2 Fleet Instance Change` event. ``` { "source": ["aws.ec2fleet"], "detail-type": ["EC2 Fleet Instance Change"] } ``` To add the event pattern, you can either use a template by choosing **Event pattern form**, or specify your own pattern by choosing **Custom pattern (JSON editor)**, as follows: 1. To use a template to create the event pattern, do the following: 1. Choose **Event pattern form**. 1. For **Event source**, choose **AWS services**. 1. For **AWS Service**, choose **EC2 Fleet**. 1. For **Event type**, choose **EC2 Fleet Instance Change**. 1. To customize the template, choose **Edit pattern** and make your changes to match the example event pattern. 1. (Alternative) To specify a custom event pattern, do the following: 1. Choose **Custom pattern (JSON editor)**. 1. In the **Event pattern** box, add the event pattern for this example. 1. Choose **Next**. 1. For **Select target(s)**, do the following: 1. For **Target types**, choose **AWS service**. 1. For **Select a target**, choose **SNS topic** to send an email, text message, or mobile push notification when the event occurs. 1. For **Topic**, choose an existing topic. You first need to create an Amazon SNS topic using the Amazon SNS console. For more information, see [Using Amazon SNS for application-to-person (A2P) messaging](https://docs.aws.amazon.com//sns/latest/dg/sns-user-notifications.html) in the *Amazon Simple Notification Service Developer Guide*. 1. (Optional) Under **Additional settings**, you can optionally configure additional settings. For more information, see [Creating Amazon EventBridge rules that react to events](https://docs.aws.amazon.com//eventbridge/latest/userguide/eb-create-rule.html) (step 16) in the *Amazon EventBridge User Guide*. 1. Choose **Next**. 1. (Optional) For **Tags**, you can optionally assign one or more tags to your rule, and then choose **Next**. 1. For **Review and create**, do the following: 1. Review the details of the rule and modify them as necessary. 1. Choose **Create rule**. For more information, see [Amazon EventBridge rules](https://docs.aws.amazon.com/eventbridge/latest/userguide/eb-rules.html) and [Amazon EventBridge event patterns](https://docs.aws.amazon.com/eventbridge/latest/userguide/eb-event-patterns.html) in the *Amazon EventBridge User Guide* ### Create an EventBridge rule to trigger a Lambda function The following example creates an EventBridge rule to trigger a Lambda function every time that Amazon EC2 emits an EC2 Fleet instance change notification for when an instance is launched. The signal in this example is emitted as an `EC2 Fleet Instance Change` event, sub-type `launched`, which triggers the action defined by the rule. Before creating the EventBridge rule, you must create the Lambda function. **To create the Lambda function to use in the EventBridge rule** 1. Open the AWS Lambda console at [https://console.aws.amazon.com/lambda/](https://console.aws.amazon.com/lambda/). 1. Choose **Create function**. 1. Enter a name for your function, configure the code, and then choose **Create function**. For more information, see [Create your first Lambda function](https://docs.aws.amazon.com/lambda/latest/dg/getting-started.html) in the *AWS Lambda Developer Guide*. **To create an EventBridge rule to trigger a Lambda function when an instance in an EC2 Fleet changes state** 1. Open the Amazon EventBridge console at [https://console.aws.amazon.com/events/](https://console.aws.amazon.com/events/). 1. Choose **Create rule**. 1. For **Define rule detail**, do the following: 1. Enter a **Name** for the rule, and, optionally, a description. A rule can't have the same name as another rule in the same Region and on the same event bus. 1. For **Event bus**, choose **default**. When an AWS service in your account generates an event, it always goes to your account's default event bus. 1. For **Rule type**, choose **Rule with an event pattern**. 1. Choose **Next**. 1. For **Build event pattern**, do the following: 1. For **Event source**, choose **AWS events or EventBridge partner events**. 1. For **Event pattern**, for this example you’ll specify the following event pattern to match the `EC2 Fleet Instance Change` event and `launched` sub-type. ``` { "source": ["aws.ec2fleet"], "detail-type": ["EC2 Fleet Instance Change"], "detail": { "sub-type": ["launched"] } ``` To add the event pattern, you can either use a template by choosing **Event pattern form**, or specify your own pattern by choosing **Custom pattern (JSON editor)**, as follows: 1. To use a template to create the event pattern, do the following: 1. Choose **Event pattern form**. 1. For **Event source**, choose **AWS services**. 1. For **AWS Service**, choose **EC2 Fleet**. 1. For **Event type**, choose **EC2 Fleet Instance Change**. 1. Choose **Edit pattern**, and add `"detail": {"sub-type": ["launched"]` to match the example event pattern. For proper JSON format, insert a comma (`,`) after the preceding square bracket (`]`). 1. (Alternative) To specify a custom event pattern, do the following: 1. Choose **Custom pattern (JSON editor)**. 1. In the **Event pattern** box, add the event pattern for this example. 1. Choose **Next**. 1. For **Select target(s)**, do the following: 1. For **Target types**, choose **AWS service**. 1. For **Select a target**, choose **SNS topic** to send an email, text message, or mobile push notification when the event occurs. 1. For **Topic**, choose **Lambda function**, and for **Function**, choose the function that you created to respond when the event occurs. 1. (Optional) Under **Additional settings**, you can optionally configure additional settings. For more information, see [Creating Amazon EventBridge rules that react to events](https://docs.aws.amazon.com//eventbridge/latest/userguide/eb-create-rule.html) (step 16) in the *Amazon EventBridge User Guide*. 1. Choose **Next**. 1. (Optional) For **Tags**, you can optionally assign one or more tags to your rule, and then choose **Next**. 1. For **Review and create**, do the following: 1. Review the details of the rule and modify them as necessary. 1. Choose **Create rule**. For a tutorial on how to create a Lambda function and an EventBridge rule that runs the Lambda function, see [Tutorial: Log the State of an Amazon EC2 Instance Using EventBridge](https://docs.aws.amazon.com/eventbridge/latest/userguide/eb-log-ec2-instance-state.html) in the *AWS Lambda Developer Guide*. ## EC2 Fleet event types There are five EC2 Fleet event types. For each event type, there are several sub-types. **Topics** + [ ### EC2 Fleet State Change ](#ec2-fleet-state-change) + [ ### EC2 Fleet Spot Instance Request Change ](#ec2-fleet-spot-instance-request-change) + [ ### EC2 Fleet Instance Change ](#ec2-fleet-instance-change) + [ ### EC2 Fleet Information ](#ec2-fleet-info) + [ ### EC2 Fleet Error ](#ec2-fleet-config-not-valid) ### EC2 Fleet State Change EC2 Fleet sends an `EC2 Fleet State Change` event to Amazon EventBridge when an EC2 Fleet changes state. The following is example data for this event. ``` { "version": "0", "id": "715ed6b3-b8fc-27fe-fad6-528c7b8bf8a2", "detail-type": "EC2 Fleet State Change", "source": "aws.ec2fleet", "account": "123456789012", "time": "2020-11-09T09:00:20Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:fleet/fleet-598fb973-87b7-422d-be4d-6b0809bfff0a" ], "detail": { "sub-type": "active" } } ``` The possible values for `sub-type` are: `active` The EC2 Fleet request has been validated and Amazon EC2 is attempting to maintain the target number of running instances. `deleted` The EC2 Fleet request is deleted and has no running instances. The EC2 Fleet will be deleted two days after its instances are terminated. `deleted_running` The EC2 Fleet request is deleted and does not launch additional instances. Its existing instances continue to run until they are interrupted or terminated. The request remains in this state until all instances are interrupted or terminated. `deleted_terminating` The EC2 Fleet request is deleted and its instances are terminating. The request remains in this state until all instances are terminated. `expired` The EC2 Fleet request has expired. If the request was created with `TerminateInstancesWithExpiration` set, a subsequent `terminated` event indicates that the instances are terminated. `modify_in_progress` The EC2 Fleet request is being modified. The request remains in this state until the modification is fully processed. `modify_succeeded` The EC2 Fleet request was modified. `submitted` The EC2 Fleet request is being evaluated and Amazon EC2 is preparing to launch the target number of instances. `progress` The EC2 Fleet request is in the process of being fulfilled. ### EC2 Fleet Spot Instance Request Change EC2 Fleet sends an `EC2 Fleet Spot Instance Request Change` event to Amazon EventBridge when a Spot Instance request in the fleet changes state. The following is example data for this event. ``` { "version": "0", "id": "19331f74-bf4b-a3dd-0f1b-ddb1422032b9", "detail-type": "EC2 Fleet Spot Instance Request Change", "source": "aws.ec2fleet", "account": "123456789012", "time": "2020-11-09T09:00:05Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:fleet/fleet-83fd4e48-552a-40ef-9532-82a3acca5f10" ], "detail": { "spot-instance-request-id": "sir-rmqske6h", "description": "SpotInstanceRequestId sir-rmqske6h, PreviousState: cancelled_running", "sub-type": "cancelled" } } ``` The possible values for `sub-type` are: `active` The Spot Instance request is fulfilled and has an associated Spot Instance. `cancelled` You cancelled the Spot Instance request, or the Spot Instance request expired. `disabled` You stopped the Spot Instance. `submitted` The Spot Instance request is submitted. ### EC2 Fleet Instance Change EC2 Fleet sends an `EC2 Fleet Instance Change` event to Amazon EventBridge when an instance in the fleet changes state. The following is example data for this event. ``` { "version": "0", "id": "542ce428-c8f1-0608-c015-e8ed6522c5bc", "detail-type": "EC2 Fleet Instance Change", "source": "aws.ec2fleet", "account": "123456789012", "time": "2020-11-09T09:00:23Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:fleet/fleet-598fb973-87b7-422d-be4d-6b0809bfff0a" ], "detail": { "instance-id": "i-0c594155dd5ff1829", "description": "{\"instanceType\":\"c5.large\",\"image\":\"ami-6057e21a\",\"productDescription\":\"Linux/UNIX\",\"availabilityZone\":\"us-east-1d\"}", "sub-type": "launched" } } ``` The possible values for `sub-type` are: `launched` A new instance was launched. `terminated` The instance was terminated. `termination_notified` An instance termination notification was sent when a Spot Instance was terminated by Amazon EC2 during scale-down, when the target capacity of the fleet was modified down, for example, from a target capacity of 4 to a target capacity of 3. ### EC2 Fleet Information EC2 Fleet sends an `EC2 Fleet Information` event to Amazon EventBridge when there is an error during fulfillment. The information event does not block the fleet from attempting to fulfil its target capacity. The following is example data for this event. ``` { "version": "0", "id": "76529817-d605-4571-7224-d36cc1b2c0c4", "detail-type": "EC2 Fleet Information", "source": "aws.ec2fleet", "account": "123456789012", "time": "2020-11-09T08:17:07Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:fleet/fleet-8becf5fe-bb9e-415d-8f54-3fa5a8628b91" ], "detail": { "description": "c4.xlarge, ami-0947d2ba12ee1ff75, Linux/UNIX, us-east-1a, Spot price in either SpotFleetRequestConfigData or SpotFleetLaunchSpecification or LaunchTemplate or LaunchTemplateOverrides is less than Spot market price $0.0619", "sub-type": "launchSpecUnusable" } } ``` The possible values for `sub-type` are: `fleetProgressHalted` The price in every launch specification is not valid because it is below the Spot price (all the launch specifications have produced `launchSpecUnusable` events). A launch specification might become valid if the Spot price changes. `launchSpecTemporarilyBlacklisted` The configuration is not valid and several attempts to launch instances have failed. For more information, see the description of the event. `launchSpecUnusable` The price in a launch specification is not valid because it is below the Spot price. `registerWithLoadBalancersFailed` An attempt to register instances with load balancers failed. For more information, see the description of the event. ### EC2 Fleet Error EC2 Fleet sends an `EC2 Fleet Error` event to Amazon EventBridge when there is an error during fulfillment. The error event blocks the fleet from attempting to fulfil its target capacity. The following is example data for this event. ``` { "version": "0", "id": "69849a22-6d0f-d4ce-602b-b47c1c98240e", "detail-type": "EC2 Fleet Error", "source": "aws.ec2fleet", "account": "123456789012", "time": "2020-10-07T01:44:24Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:fleet/fleet-9bb19bc6-60d3-4fd2-ae47-d33e68eafa08" ], "detail": { "description": "m3.large, ami-00068cd7555f543d5, Linux/UNIX: IPv6 is not supported for the instance type 'm3.large'. ", "sub-type": "spotFleetRequestConfigurationInvalid" } } ``` The possible values for `sub-type` are: `iamFleetRoleInvalid` The EC2 Fleet does not have the required permissions to either launch or terminate an instance. `allLaunchSpecsTemporarilyBlacklisted` None of the configurations are valid, and several attempts to launch instances have failed. For more information, see the description of the event. `spotInstanceCountLimitExceeded` You’ve reached the limit on the number of Spot Instances that you can launch. `spotFleetRequestConfigurationInvalid` The configuration is not valid. For more information, see the description of the event. ## Spot Fleet event types There are five Spot Fleet event types. For each event type, there are several sub-types. **Topics** + [ ### EC2 Spot Fleet State Change ](#spot-fleet-state-change) + [ ### EC2 Spot Fleet Spot Instance Request Change ](#spot-fleet-spot-instance-request-change) + [ ### EC2 Spot Fleet Instance Change ](#spot-fleet-instance-change) + [ ### EC2 Spot Fleet Information ](#spot-fleet-info) + [ ### EC2 Spot Fleet Error ](#spot-fleet-config-not-valid) ### EC2 Spot Fleet State Change Spot Fleet sends an `EC2 Spot Fleet State Change` event to Amazon EventBridge when a Spot Fleet changes state. The following is example data for this event. ``` { "version": "0", "id": "d1af1091-6cc3-2e24-203a-3b870e455d5b", "detail-type": "EC2 Spot Fleet State Change", "source": "aws.ec2spotfleet", "account": "123456789012", "time": "2020-11-09T08:57:06Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:spot-fleet-request/sfr-4b6d274d-0cea-4b2c-b3be-9dc627ad1f55" ], "detail": { "sub-type": "submitted" } } ``` The possible values for `sub-type` are: `active` The Spot Fleet request has been validated and Amazon EC2 is attempting to maintain the target number of running instances. `cancelled` The Spot Fleet request is canceled and has no running instances. The Spot Fleet will be deleted two days after its instances are terminated. `cancelled_running` The Spot Fleet request is canceled and does not launch additional instances. Its existing instances continue to run until they are interrupted or terminated. The request remains in this state until all instances are interrupted or terminated. `cancelled_terminating` The Spot Fleet request is canceled and its instances are terminating. The request remains in this state until all instances are terminated. `expired` The Spot Fleet request has expired. If the request was created with `TerminateInstancesWithExpiration` set, a subsequent `terminated` event indicates that the instances are terminated. `modify_in_progress` The Spot Fleet request is being modified. The request remains in this state until the modification is fully processed. `modify_succeeded` The Spot Fleet request was modified. `submitted` The Spot Fleet request is being evaluated and Amazon EC2 is preparing to launch the target number of instances. `progress` The Spot Fleet request is in the process of being fulfilled. ### EC2 Spot Fleet Spot Instance Request Change Spot Fleet sends an `EC2 Spot Fleet Spot Instance Request Change` event to Amazon EventBridge when a Spot Instance request in the fleet changes state. The following is example data for this event. ``` { "version": "0", "id": "cd141ef0-14af-d670-a71d-fe46e9971bd2", "detail-type": "EC2 Spot Fleet Spot Instance Request Change", "source": "aws.ec2spotfleet", "account": "123456789012", "time": "2020-11-09T08:53:21Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:spot-fleet-request/sfr-a98d2133-941a-47dc-8b03-0f94c6852ad1" ], "detail": { "spot-instance-request-id": "sir-a2w9gc5h", "description": "SpotInstanceRequestId sir-a2w9gc5h, PreviousState: cancelled_running", "sub-type": "cancelled" } } ``` The possible values for `sub-type` are: `active` The Spot Instance request is fulfilled and has an associated Spot Instance. `cancelled` You cancelled the Spot Instance request, or the Spot Instance request expired. `disabled` You stopped the Spot Instance. `submitted` The Spot Instance request is submitted. ### EC2 Spot Fleet Instance Change Spot Fleet sends an `EC2 Spot Fleet Instance Change` event to Amazon EventBridge when an instance in the fleet changes state. The following is example data for this event. ``` { "version": "0", "id": "11591686-5bd7-bbaa-eb40-d46529c2710f", "detail-type": "EC2 Spot Fleet Instance Change", "source": "aws.ec2spotfleet", "account": "123456789012", "time": "2020-11-09T07:25:02Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:spot-fleet-request/sfr-c8a764a4-bedc-4b62-af9c-0095e6e3ba61" ], "detail": { "instance-id": "i-08b90df1e09c30c9b", "description": "{\"instanceType\":\"r4.2xlarge\",\"image\":\"ami-032930428bf1abbff\",\"productDescription\":\"Linux/UNIX\",\"availabilityZone\":\"us-east-1a\"}", "sub-type": "launched" } } ``` The possible values for `sub-type` are: `launched` A new instance was launched. `terminated` The instance was terminated. `termination_notified` An instance termination notification was sent when a Spot Instance was terminated by Amazon EC2 during scale-down, when the target capacity of the fleet was modified down, for example, from a target capacity of 4 to a target capacity of 3. ### EC2 Spot Fleet Information Spot Fleet sends an `EC2 Spot Fleet Information` event to Amazon EventBridge when there is an error during fulfillment. The information event does not block the fleet from attempting to fulfil its target capacity. The following is example data for this event. ``` { "version": "0", "id": "73a60f70-3409-a66c-635c-7f66c5f5b669", "detail-type": "EC2 Spot Fleet Information", "source": "aws.ec2spotfleet", "account": "123456789012", "time": "2020-11-08T20:56:12Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:spot-fleet-request/sfr-2531ea06-af18-4647-8757-7d69c94971b1" ], "detail": { "description": "r3.8xlarge, ami-032930428bf1abbff, Linux/UNIX, us-east-1a, Spot bid price is less than Spot market price $0.5291", "sub-type": "launchSpecUnusable" } } ``` The possible values for `sub-type` are: `fleetProgressHalted` The price in every launch specification is not valid because it is below the Spot price (all the launch specifications have produced `launchSpecUnusable` events). A launch specification might become valid if the Spot price changes. `launchSpecTemporarilyBlacklisted` The configuration is not valid and several attempts to launch instances have failed. For more information, see the description of the event. `launchSpecUnusable` The price in a launch specification is not valid because it is below the Spot price. `registerWithLoadBalancersFailed` An attempt to register instances with load balancers failed. For more information, see the description of the event. ### EC2 Spot Fleet Error Spot Fleet sends an `EC2 Spot Fleet Error` event to Amazon EventBridge when there is an error during fulfillment. The error event blocks the fleet from attempting to fulfil its target capacity. The following is example data for this event. ``` { "version": "0", "id": "10adc4e7-675c-643e-125c-5bfa1b1ba5d2", "detail-type": "EC2 Spot Fleet Error", "source": "aws.ec2spotfleet", "account": "123456789012", "time": "2020-11-09T06:56:07Z", "region": "us-east-1", "resources": [ "arn:aws:ec2:us-east-1:123456789012:spot-fleet-request/sfr-38725d30-25f1-4f30-83ce-2907c56dba17" ], "detail": { "description": "r4.2xlarge, ami-032930428bf1abbff, Linux/UNIX: The associatePublicIPAddress parameter can only be specified for the network interface with DeviceIndex 0. ", "sub-type": "spotFleetRequestConfigurationInvalid" } } ``` The possible values for `sub-type` are: `iamFleetRoleInvalid` The Spot Fleet does not have the required permissions to either launch or terminate an instance. `allLaunchSpecsTemporarilyBlacklisted` None of the configurations are valid, and several attempts to launch instances have failed. For more information, see the description of the event. `spotInstanceCountLimitExceeded` You’ve reached the limit on the number of Spot Instances that you can launch. `spotFleetRequestConfigurationInvalid` The configuration is not valid. For more information, see the description of the event. # Tutorials for EC2 Fleet There are different ways to configure an EC2 Fleet. The configuration you choose depends on your specific use case. The following tutorials cover some of the possible use cases and provide the tasks required to implement them. | Use case | Link to tutorial | | --- | --- | | **Use instance weighting to manage the availability and performance of your EC2 Fleet.** With instance weighting, you assign a weight to each instance type in your EC2 Fleet to represent their compute capacity and performance relative to each other. Based on the weights, the fleet can use any combination of the specified instance types, as long as it can fulfil the desired target capacity. | [Tutorial: Configure EC2 Fleet to use instance weighting](ec2-fleet-instance-weighting-walkthrough.md) | | **Use On-Demand capacity to ensure availability during peak periods, but benefit from additional Spot capacity at a lower cost.** Configure your EC2 Fleet to use On-Demand Instances as the primary capacity to ensure available capacity during peak periods. In addition, allocate some capacity to Spot Instances to benefit from discounted pricing, while keeping in mind that Spot Instances can be interrupted if Amazon EC2 needs the capacity back. | [Tutorial: Configure EC2 Fleet to use On-Demand Instances as the primary capacity](ec2-fleet-on-demand-walkthrough.md) | | **Use Capacity Reservations to reserve compute capacity for your On-Demand Instances.** Configure your EC2 Fleet to use `targeted` Capacity Reservations first when launching On-Demand Instances. If you have strict capacity requirements, and are running business-critical workloads that require a certain level of long or short-term capacity assurance, we recommend that you create a Capacity Reservation to ensure that you always have access to Amazon EC2 capacity when you need it, for as long as you need it. | [Tutorial: Configure EC2 Fleet to launch On-Demand Instances using targeted Capacity Reservations](ec2-fleet-launch-on-demand-instances-using-targeted-capacity-reservations-walkthrough.md) | | **Use Capacity Blocks to reserve highly sought-after GPU instances for your ML workloads.** Configure your EC2 Fleet to launch instances into Capacity Blocks. | [Tutorial: Configure your EC2 Fleet to launch instances into Capacity Blocks](ec2-fleet-launch-instances-capacity-blocks-walkthrough.md) | | **Use Interruptible Capacity Reservations to temporarily repurpose idle capacity across your AWS Organization and save costs.** Configure your EC2 Fleet to launch instances into Interruptible Capacity Reservations, which represent spare capacity within your AWS organization. Capacity owners can reclaim the capacity at any time. Once reclaimed, EC2 terminates the instances after a 2-minute notice. | [Tutorial: Configure your EC2 Fleet to launch instances into Interruptible Capacity Reservations](ec2-fleet-launch-instances-interruptible-cr-walkthrough.md) | # Tutorial: Configure EC2 Fleet to use instance weighting This tutorial uses a fictitious company called Example Corp to illustrate the process of requesting an EC2 Fleet using instance weighting. ## Objective Example Corp, a pharmaceutical company, wants to use the computational power of Amazon EC2 for screening chemical compounds that might be used to fight cancer. ## Planning Example Corp first reviews [Spot Best Practices](https://aws.amazon.com/ec2/spot/getting-started/#bestpractices). Next, Example Corp determines the requirements for their EC2 Fleet. **Instance types** Example Corp has a compute- and memory-intensive application that performs best with at least 60 GB of memory and eight virtual CPUs (vCPUs). They want to maximize these resources for the application at the lowest possible price. Example Corp decides that any of the following EC2 instance types would meet their needs: | Instance type | Memory (GiB) | vCPUs | | --- | --- | --- | | r3.2xlarge | 61 | 8 | | r3.4xlarge | 122 | 16 | | r3.8xlarge | 244 | 32 | **Target capacity in units** With instance weighting, target capacity can equal a number of instances (the default) or a combination of factors such as cores (vCPUs), memory (GiBs), and storage (GBs). By considering the base for their application (60 GB of RAM and eight vCPUs) as one unit, Example Corp decides that 20 times this amount would meet their needs. So the company sets the target capacity of their EC2 Fleet request to 20 units. **Instance weights** After determining the target capacity, Example Corp calculates instance weights. To calculate the instance weight for each instance type, they determine the units of each instance type that are required to reach the target capacity as follows: + r3.2xlarge (61.0 GB, 8 vCPUs) = 1 unit of 20 + r3.4xlarge (122.0 GB, 16 vCPUs) = 2 units of 20 + r3.8xlarge (244.0 GB, 32 vCPUs) = 4 units of 20 Therefore, Example Corp assigns instance weights of 1, 2, and 4 to the respective launch configurations in their EC2 Fleet request. **Price per unit hour** Example Corp uses the [On-Demand price](https://aws.amazon.com/ec2/pricing/) per instance hour as a starting point for their price. They could also use recent Spot prices, or a combination of the two. To calculate the price per unit hour, they divide their starting price per instance hour by the weight. For example: | Instance type | On-Demand price | Instance weight | Price per unit hour | | --- | --- | --- | --- | | r3.2xLarge | \$10.7 | 1 | \$10.7 | | r3.4xLarge | \$11.4 | 2 | \$10.7 | | r3.8xLarge | \$12.8 | 4 | \$10.7 | Example Corp could use a global price per unit hour of \$10.7 and be competitive for all three instance types. They could also use a global price per unit hour of \$10.7 and a specific price per unit hour of \$10.9 in the `r3.8xlarge` launch specification. ## Verify permissions Before creating an EC2 Fleet, Example Corp verifies that it has an IAM role with the required permissions. For more information, see [EC2 Fleet prerequisites](ec2-fleet-prerequisites.md). ## Create a launch template Next, Example Corp creates a launch template. The launch template ID is used in the following step. For more information, see [Create an Amazon EC2 launch template](create-launch-template.md). ## Create the EC2 Fleet Example Corp creates a file, `config.json`, with the following configuration for its EC2 Fleet. In the following example, replace the resource identifiers with your own resource identifiers. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-07b3bc7625cdab851", "Version": "1" }, "Overrides": [ { "InstanceType": "r3.2xlarge", "SubnetId": "subnet-482e4972", "WeightedCapacity": 1 }, { "InstanceType": "r3.4xlarge", "SubnetId": "subnet-482e4972", "WeightedCapacity": 2 }, { "InstanceType": "r3.8xlarge", "MaxPrice": "0.90", "SubnetId": "subnet-482e4972", "WeightedCapacity": 4 } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" } } ``` Example Corp creates the EC2 Fleet using the following [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command. ``` aws ec2 create-fleet --cli-input-json file://config.json ``` For more information, see [Create an EC2 Fleet](create-ec2-fleet.md). ## Fulfillment The allocation strategy determines which Spot capacity pools your Spot Instances come from. With the `lowest-price` strategy (which is the default strategy), the Spot Instances come from the pool with the lowest price per unit at the time of fulfillment. To provide 20 units of capacity, the EC2 Fleet launches either 20 `r3.2xlarge` instances (20 divided by 1), 10 `r3.4xlarge` instances (20 divided by 2), or 5 `r3.8xlarge` instances (20 divided by 4). If Example Corp used the `diversified` strategy, the Spot Instances would come from all three pools. The EC2 Fleet would launch 6 `r3.2xlarge` instances (which provide 6 units), 3 `r3.4xlarge` instances (which provide 6 units), and 2 `r3.8xlarge` instances (which provide 8 units), for a total of 20 units. # Tutorial: Configure EC2 Fleet to use On-Demand Instances as the primary capacity This tutorial uses a fictitious company called ABC Online to illustrate the process of requesting an EC2 Fleet with On-Demand as the primary capacity, and Spot capacity if available. ## Objective ABC Online, a restaurant delivery company, aims to provision Amazon EC2 capacity across EC2 instance types and purchasing options to achieve their desired scale, performance, and cost. ## Plan ABC Online requires a fixed capacity to handle peak periods, but would like to benefit from additional capacity at a lower cost. The company determines the following requirements for their EC2 Fleet: + On-Demand Instance capacity – ABC Online requires 15 On-Demand Instances to ensure that they can accommodate traffic at peak periods. + Spot Instance capacity – To enhance performance, but at a lower price, ABC Online plans to provision 5 Spot Instances. ## Verify permissions Before creating an EC2 Fleet, ABC Online verifies that it has an IAM role with the required permissions. For more information, see [EC2 Fleet prerequisites](ec2-fleet-prerequisites.md). ## Create a launch template Next, ABC Online creates a launch template. The launch template ID is used in the following step. For more information, see [Create an Amazon EC2 launch template](create-launch-template.md). ## Create the EC2 Fleet ABC Online creates a file, `config.json`, with the following configuration for its EC2 Fleet. In the following example, replace the resource identifiers with your own resource identifiers. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-07b3bc7625cdab851", "Version": "2" } } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "OnDemandTargetCapacity":15, "DefaultTargetCapacityType": "spot" } } ``` ABC Online creates the EC2 Fleet using the following [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command. ``` aws ec2 create-fleet --cli-input-json file://config.json ``` For more information, see [Create an EC2 Fleet](create-ec2-fleet.md). ## Fulfillment The allocation strategy determines that the On-Demand capacity is always fulfilled, while the balance of the target capacity is fulfilled as Spot if there is available capacity. # Tutorial: Configure EC2 Fleet to launch On-Demand Instances using targeted Capacity Reservations This tutorial walks you through all the steps that you must perform so that your EC2 Fleet launches On-Demand Instances into `targeted` Capacity Reservations. You will learn how to configure a fleet to use `targeted` On-Demand Capacity Reservations first when launching On-Demand Instances. You will also learn how to configure the fleet so that, when the total On-Demand target capacity exceeds the number of available unused Capacity Reservations, the fleet uses the specified allocation strategy for selecting the instance pools in which to launch the remaining target capacity. **EC2 Fleet configuration** In this tutorial, the fleet is configured as follows: + Target capacity: 10 On-Demand Instances + Total unused `targeted` Capacity Reservations: 6 (less than the fleet's On-Demand target capacity of 10 On-Demand Instances) + Number of Capacity Reservation pools: 2 (`us-east-1a` and `us-east-1b`) + Number of Capacity Reservations per pool: 3 + On-Demand allocation strategy: `lowest-price` (When the number of unused Capacity Reservations is less than the On-Demand target capacity, the fleet determines the pools in which to launch the remaining On-Demand capacity based on the On-Demand allocation strategy.) Note that you can also use the `prioritized` allocation strategy instead of the `lowest-price` allocation strategy. **To launch On-Demand Instances into `targeted` Capacity Reservations, you must perform a number of steps, as follows:** + [ ## Step 1: Create Capacity Reservations ](#ec2-fleet-odcr-step1) + [ ## Step 2: Create a Capacity Reservation resource group ](#ec2-fleet-odcr-step2) + [ ## Step 3: Add the Capacity Reservations to the Capacity Reservation resource group ](#ec2-fleet-odcr-step3) + [ ## (Optional) Step 4: View the Capacity Reservations in the resource group ](#ec2-fleet-odcr-step4) + [ ## Step 5: Create a launch template that specifies that the Capacity Reservation targets a specific resource group ](#ec2-fleet-odcr-step5) + [ ## (Optional) Step 6: Describe the launch template ](#ec2-fleet-odcr-step6) + [ ## Step 7: Create an EC2 Fleet ](#ec2-fleet-odcr-step7) + [ ## (Optional) Step 8: View the number of remaining unused Capacity Reservations ](#ec2-fleet-odcr-step8) ## Step 1: Create Capacity Reservations Use the [create-capacity-reservation](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-capacity-reservation.html) command to create the Capacity Reservations, three for `us-east-1a` and another three for `us-east-1b`. Except for the Availability Zone, the other attributes of the Capacity Reservations are identical. **3 Capacity Reservations in `us-east-1a`** ``` aws ec2 create-capacity-reservation \ --availability-zone us-east-1a \ --instance-type c5.xlarge \ --instance-platform Linux/UNIX \ --instance-count 3 \ --instance-match-criteria targeted ``` Example of resulting Capacity Reservation ID ``` cr-1234567890abcdef1 ``` **3 Capacity Reservations in `us-east-1b`** ``` aws ec2 create-capacity-reservation \ --availability-zone us-east-1b \ --instance-type c5.xlarge \ --instance-platform Linux/UNIX \ --instance-count 3 \ --instance-match-criteria targeted ``` Example of resulting Capacity Reservation ID ``` cr-54321abcdef567890 ``` ## Step 2: Create a Capacity Reservation resource group Use the `resource-groups` service and the [create-group](https://docs.aws.amazon.com/cli/latest/reference/resource-groups/create-group.html) command to create a Capacity Reservation resource group. In this example, the resource group is named `my-cr-group`. For information about why you must create a resource group, see [Use Capacity Reservations to reserve On-Demand capacity in EC2 Fleet](ec2-fleet-on-demand-capacity-reservations.md). ``` aws resource-groups create-group \ --name my-cr-group \ --configuration '{"Type":"AWS::EC2::CapacityReservationPool"}' '{"Type":"AWS::ResourceGroups::Generic", "Parameters": [{"Name": "allowed-resource-types", "Values": ["AWS::EC2::CapacityReservation"]}]}' ``` ## Step 3: Add the Capacity Reservations to the Capacity Reservation resource group Use the `resource-groups` service and the [group-resources](https://docs.aws.amazon.com/cli/latest/reference/resource-groups/group-resources.html) command to add the Capacity Reservations that you created in Step 1 to the Capacity Reservations resource group. Note that you must reference the On-Demand Capacity Reservations by their ARNs. ``` aws resource-groups group-resources \ --group my-cr-group \ --resource-arns \ arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-1234567890abcdef1 \ arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-54321abcdef567890 ``` Example output ``` { "Failed": [], "Succeeded": [ "arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-1234567890abcdef1", "arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-54321abcdef567890" ] } ``` ## (Optional) Step 4: View the Capacity Reservations in the resource group Use the `resource-groups` service and the [list-group-resources](https://docs.aws.amazon.com/cli/latest/reference/resource-groups/list-group-resources.html) command to optionally describe the resource group to view its Capacity Reservations. ``` aws resource-groups list-group-resources --group my-cr-group ``` Example output ``` { "ResourceIdentifiers": [ { "ResourceType": "AWS::EC2::CapacityReservation", "ResourceArn": "arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-1234567890abcdef1" }, { "ResourceType": "AWS::EC2::CapacityReservation", "ResourceArn": "arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-54321abcdef567890" } ] } ``` ## Step 5: Create a launch template that specifies that the Capacity Reservation targets a specific resource group Use the [create-launch-template](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-launch-template.html) command to create a launch template in which to specify the Capacity Reservations to use. In this example, the fleet will use `targeted` Capacity Reservations, which have been added to a resource group. Therefore, the launch template data specifies that the Capacity Reservation targets a specific resource group. In this example, the launch template is named `my-launch-template`. ``` aws ec2 create-launch-template \ --launch-template-name my-launch-template \ --launch-template-data \ '{"ImageId": "ami-0123456789example", "CapacityReservationSpecification": {"CapacityReservationTarget": { "CapacityReservationResourceGroupArn": "arn:aws:resource-groups:us-east-1:123456789012:group/my-cr-group" } } }' ``` ## (Optional) Step 6: Describe the launch template Use the [describe-launch-template-versions](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-launch-template-versions.html) command to optionally describe the launch template to view its configuration. ``` aws ec2 describe-launch-template-versions --launch-template-name my-launch-template ``` Example output ``` { "LaunchTemplateVersions": [ { "LaunchTemplateId": "lt-01234567890example", "LaunchTemplateName": "my-launch-template", "VersionNumber": 1, "CreateTime": "2021-01-19T20:50:19.000Z", "CreatedBy": "arn:aws:iam::123456789012:user/Admin", "DefaultVersion": true, "LaunchTemplateData": { "ImageId": "ami-0947d2ba12ee1ff75", "CapacityReservationSpecification": { "CapacityReservationTarget": { "CapacityReservationResourceGroupArn": "arn:aws:resource-groups:us-east-1:123456789012:group/my-cr-group" } } } } ] } ``` ## Step 7: Create an EC2 Fleet Create an EC2 Fleet that specifies the configuration information for the instances that it will launch. The following EC2 Fleet configuration shows only the pertinent configurations for this example. The launch template `my-launch-template` is the launch template you created in Step 5. There are two instance pools, each with the same instance type (`c5.xlarge`), but with different Availability Zones (`us-east-1a` and `us-east-1b`). The price of the instance pools is the same because pricing is defined for the Region, not per Availability Zone. The total target capacity is 10, and the default target capacity type is `on-demand`. The On-Demand allocation strategy is `lowest-price`. The usage strategy for Capacity Reservations is `use-capacity-reservations-first`. **Note** The fleet type must be `instant`. Other fleet types do not support `use-capacity-reservations-first`. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "c5.xlarge", "AvailabilityZone": "us-east-1a" }, { "InstanceType": "c5.xlarge", "AvailabilityZone": "us-east-1b" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 10, "DefaultTargetCapacityType": "on-demand" }, "OnDemandOptions": { "AllocationStrategy": "lowest-price", "CapacityReservationOptions": { "UsageStrategy": "use-capacity-reservations-first" } }, "Type": "instant" } ``` After you create the `instant` fleet using the preceding configuration, the following 10 instances are launched to meet the target capacity: + The Capacity Reservations are used first to launch 6 On-Demand Instances as follows: + 3 On-Demand Instances are launched into the 3 `c5.xlarge` `targeted` Capacity Reservations in `us-east-1a` + 3 On-Demand Instances are launched into the 3 `c5.xlarge` `targeted` Capacity Reservations in `us-east-1b` + To meet the target capacity, 4 additional On-Demand Instances are launched into regular On-Demand capacity according to the On-Demand allocation strategy, which is `lowest-price` in this example. However, because the pools are the same price (because price is per Region and not per Availability Zone), the fleet launches the remaining 4 On-Demand Instances into either of the pools. ## (Optional) Step 8: View the number of remaining unused Capacity Reservations After the fleet is launched, you can optionally run [describe-capacity-reservations](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-capacity-reservations.html) to see how many unused Capacity Reservations are remaining. In this example, you should see the following response, which shows that all of the Capacity Reservations in all of the pools were used. ``` { "CapacityReservationId": "cr-111", "InstanceType": "c5.xlarge", "AvailableInstanceCount": 0 } { "CapacityReservationId": "cr-222", "InstanceType": "c5.xlarge", "AvailableInstanceCount": 0 } ``` # Tutorial: Configure your EC2 Fleet to launch instances into Capacity Blocks This tutorial walks you through the steps that you must perform so that your EC2 Fleet launches instances into Capacity Blocks. In most cases, the target capacity of the EC2 Fleet request should be less than or equal to the available capacity of the Capacity Block reservation that you are targeting. Target capacity requests that exceed the limits of the Capacity Block reservation will not be fulfilled. If the target capacity request exceeds the limits of your Capacity Block reservation, you will receive an `Insufficient Capacity Exception` for the capacity that exceeds the limits of your Capacity Block reservation. **Note** For Capacity Blocks, EC2 Fleet will not fallback to launching On-Demand Instances for the remainder of the desired target capacity. If EC2 Fleet is unable to fulfill the requested target capacity in an available Capacity Block reservation, EC2 Fleet will fulfill as much capacity as it can and return the instances that it was able to launch. You can repeat the call to EC2 Fleet again until all the instances are provisioned. After configuring the EC2 Fleet request, you must wait until the start date of your Capacity Block reservation. If you make requests to EC2 Fleet to launch into a Capacity Block that hasn't started yet, you will receive an `Insufficient Capacity Error`. After your Capacity Block reservation becomes active, you can make EC2 Fleet API calls and provision the instances into your Capacity Block based on the parameters you selected. Instances running in the Capacity Block continue to run until you manually stop or terminate them or until Amazon EC2 terminates the instances when the Capacity Block reservation ends. For more information about Capacity Blocks, see [Capacity Blocks for ML](ec2-capacity-blocks.md). **Considerations** + Only EC2 Fleet requests of type `instant` are supported for launching instances into Capacity Blocks. For more information, see [Configure an EC2 Fleet of type instant](instant-fleet.md). + Multiple Capacity Blocks in the same EC2 Fleet request aren't supported. + Using `OnDemandTargetCapacity` or `SpotTargetCapacity` while also setting `capacity-block` as the `DefaultTargetCapacity` isn't supported. + If `DefaultTargetCapacityType` is set to `capacity-block`, you can't provide `OnDemandOptions::CapacityReservationOptions`. An exception will occur. **To configure an EC2 Fleet to launch instances into Capacity Blocks** 1. **Create a launch template.** In the launch template, do the following: + For `InstanceMarketOptionsRequest`, set `MarketType` to `capacity-block`. + To target the Capacity Block reservation, for `CapacityReservationID`, specify the Capacity Block reservation ID. Make note of launch template name and version. You'll use this information in the next step. For more information about creating a launch template, see [Create an Amazon EC2 launch template](create-launch-template.md). 1. **Configure the EC2 Fleet.** Create a file, `config.json`, with the following configuration for your EC2 Fleet. In the following example, replace the resource identifiers with your own resource identifiers. For more information about configuring an EC2 Fleet, see [Create an EC2 Fleet](create-ec2-fleet.md). ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "CBR-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "p5.48xlarge", "AvailabilityZone": "us-east-1a" }, ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 10, "DefaultTargetCapacityType": "capacity-block" }, "Type": "instant" } ``` 1. **Launch the fleet.** Use the following [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command. ``` aws ec2 create-fleet --cli-input-json file://config.json ``` For more information, see [Create an EC2 Fleet](create-ec2-fleet.md#create-ec2-fleet-procedure). # Tutorial: Configure your EC2 Fleet to launch instances into Interruptible Capacity Reservations This tutorial walks you through the steps that you must perform so that your EC2 Fleet launches instances into Interruptible Capacity Reservations. Interruptible Capacity Reservations represent spare capacity lent to you by On-Demand Capacity Reservation owners within your AWS organization. These reservations are suitable for interruptible workloads because the capacity owner can reclaim the capacity at any time with a 2-minute interruption notice, after which EC2 terminates the instances. For more information about Interruptible Capacity Reservations, see [Interruptible Capacity Reservations](interruptible-capacity-reservations.md). The target capacity of the EC2 Fleet request should be less than or equal to the available capacity of the Interruptible Capacity Reservation that you are targeting. If the target capacity request exceeds the available capacity of your Interruptible Capacity Reservation, EC2 Fleet launches as many instances as it can and reports the launched instances in the API response. Instances running in an Interruptible Capacity Reservation continue to run until you manually stop or terminate them, or until the capacity owner reclaims the capacity. When the capacity owner reclaims the capacity, Amazon EC2 sends an Amazon EventBridge notification 2 minutes before terminating the instances. ## Considerations + Only EC2 Fleet requests of type `instant` are supported for launching instances into Interruptible Capacity Reservations. + Using `OnDemandTargetCapacity` or `SpotTargetCapacity` while also setting `reserved-capacity` as the `DefaultTargetCapacityType` is not supported. + When you specify multiple launch templates, each targeting a different Interruptible Capacity Reservation, EC2 Fleet provisions instances across all matching reservations. + For Interruptible Capacity Reservations, EC2 Fleet does not fall back to launching On-Demand Instances or Spot Instances for the remainder of the desired target capacity. ## Verify permissions Before creating an EC2 Fleet, verify that you have an IAM role with the required permissions. For more information, see [EC2 Fleet prerequisites](ec2-fleet-prerequisites.md). To launch instances into an Interruptible Capacity Reservation, you must perform the following steps: ## Step 1: Create a launch template Use the [create-launch-template](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-launch-template.html) command to create a launch template that specifies the Interruptible Capacity Reservation to target. In the launch template, set `MarketType` to `interruptible-capacity-reservation` and specify the `CapacityReservationId` of your Interruptible Capacity Reservation. Example launch template configuration: ``` { "LaunchTemplateName": "interruptible-cr-launch-template", "LaunchTemplateData": { "InstanceType": "m5.large", "ImageId": "ami-0abcdef1234567890", "CapacityReservationSpecification": { "CapacityReservationTarget": { "CapacityReservationId": "cr-0123456789abcdef0" } }, "InstanceMarketOptions": { "MarketType": "interruptible-capacity-reservation" } } } ``` Example output ``` { "LaunchTemplate": { "LaunchTemplateId": "lt-0123456789example", "LaunchTemplateName": "interruptible-cr-launch-template", "CreateTime": "2026-03-12T10:00:00.000Z", "CreatedBy": "arn:aws:iam::123456789012:user/Admin", "DefaultVersionNumber": 1, "LatestVersionNumber": 1 } } ``` For more information, see [Create an Amazon EC2 launch template](create-launch-template.md). ## Step 2: Configure the EC2 Fleet Create a configuration file for the EC2 Fleet that specifies the launch template and target capacity. The following configuration uses the launch template `interruptible-cr-launch-template` that you created in Step 1. You must specify `ReservedCapacityOptions` with `ReservationType` set to `interruptible-capacity-reservation` when using `reserved-capacity` as the `DefaultTargetCapacityType`. Create a file named `config.json` with the following content: ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "interruptible-cr-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "m5.large", "AvailabilityZone": "us-east-1a" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 10, "DefaultTargetCapacityType": "reserved-capacity" }, "ReservedCapacityOptions": { "ReservationType": ["interruptible-capacity-reservation"] }, "Type": "instant" } ``` Key configuration parameters: | Parameter | Description | | --- | --- | | DefaultTargetCapacityType | Set to reserved-capacity to indicate that instances should be launched into reserved capacity. | | ReservedCapacityOptions | Specifies the type of reserved capacity. For Interruptible Capacity Reservations, set ReservationType to ["interruptible-capacity-reservation"]. | | Type | Must be set to instant. Only instant fleets are supported for Interruptible Capacity Reservations. | ## Step 3: Launch the fleet and view results Use the [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command to create the fleet: ``` aws ec2 create-fleet \ --cli-input-json file://config.json ``` After you create the `instant` fleet using the preceding configuration, EC2 Fleet launches 10 instances into the Interruptible Capacity Reservation to meet the target capacity. **Note** If the fleet cannot fulfill the full target capacity, the response includes the instances that were launched and any errors for unfulfilled capacity. Example output ``` { "FleetId": "fleet-12345678-1234-1234-1234-123456789012", "Instances": [ { "LaunchTemplateAndOverrides": { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0123456789example", "Version": "1" }, "Overrides": { "InstanceType": "m5.large", "AvailabilityZone": "us-east-1a" } }, "Lifecycle": "interruptible-capacity-reservation", "InstanceIds": [ "i-0123456789example1", "i-0123456789example2", "i-0123456789example3", "i-0123456789example4", "i-0123456789example5", "i-0123456789example6", "i-0123456789example7", "i-0123456789example8", "i-0123456789example9", "i-0123456789example0" ], "InstanceType": "m5.large", "Platform": "Linux/UNIX" } ], "Errors": [] } ``` For more information, see [Create an EC2 Fleet](create-ec2-fleet.md). ## Clean up To stop incurring charges, terminate the instances when they are no longer needed. Note that EC2 also terminates instances launched into an Interruptible Capacity Reservation automatically when the capacity owner reclaims the capacity. ## Related resources + [Interruptible Capacity Reservations](interruptible-capacity-reservations.md) + [Reserve compute capacity with EC2 On-Demand Capacity Reservations](ec2-capacity-reservations.md) + [Work with EC2 Fleet](manage-ec2-fleet.md) + [Create an EC2 Fleet](create-ec2-fleet.md) + [Store instance launch parameters in Amazon EC2 launch templates](ec2-launch-templates.md) # Example CLI configurations for EC2 Fleet You can define your EC2 Fleet configuration in a JSON file, and then reference that file with the [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html) command to create your fleet, as follows: ``` aws ec2 create-fleet --cli-input-json file://file_name.json ``` The following examples illustrate launch configurations for various EC2 Fleet use cases. For more information about the configuration parameters, see [create-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/create-fleet.html). **Topics** + [ ## Example 1: Launch Spot Instances as the default purchasing option ](#ec2-fleet-config1) + [ ## Example 2: Launch On-Demand Instances as the default purchasing option ](#ec2-fleet-config2) + [ ## Example 3: Launch On-Demand Instances as the primary capacity ](#ec2-fleet-config3) + [ ## Example 4: Launch On-Demand Instances using multiple Capacity Reservations ](#ec2-fleet-config5) + [ ## Example 5: Launch On-Demand Instances using Capacity Reservations when the total target capacity exceeds the number of unused Capacity Reservations ](#ec2-fleet-config6) + [ ## Example 6: Launch On-Demand Instances using targeted Capacity Reservations ](#ec2-fleet-config7) + [ ## Example 7: Configure Capacity Rebalancing to launch replacement Spot Instances ](#ec2-fleet-config8) + [ ## Example 8: Launch Spot Instances in a capacity-optimized fleet ](#ec2-fleet-config9) + [ ## Example 9: Launch Spot Instances in a capacity-optimized fleet with priorities ](#ec2-fleet-config10) + [ ## Example 10: Launch Spot Instances in a price-capacity-optimized fleet ](#ec2-fleet-config11) + [ ## Example 11: Configure attribute-based instance type selection ](#ec2-fleet-config12) For more CLI examples for fleets of type `instant`, see [Configure an EC2 Fleet of type instant](instant-fleet.md). ## Example 1: Launch Spot Instances as the default purchasing option The following example specifies the minimum parameters required in an EC2 Fleet: a launch template, target capacity, and default purchasing option. The launch template is identified by its launch template ID and version number. The target capacity for the fleet is 2 instances, and the default purchasing option is `spot`, which results in the fleet launching 2 Spot Instances. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0e8c754449b27161c", "Version": "1" } } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 2, "DefaultTargetCapacityType": "spot" } } ``` ## Example 2: Launch On-Demand Instances as the default purchasing option The following example specifies the minimum parameters required in an EC2 Fleet: a launch template, target capacity, and default purchasing option. The launch template is identified by its launch template ID and version number. The target capacity for the fleet is 2 instances, and the default purchasing option is `on-demand`, which results in the fleet launching 2 On-Demand Instances. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0e8c754449b27161c", "Version": "1" } } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 2, "DefaultTargetCapacityType": "on-demand" } } ``` ## Example 3: Launch On-Demand Instances as the primary capacity The following example specifies the total target capacity of 2 instances for the fleet, and a target capacity of 1 On-Demand Instance. The default purchasing option is `spot`. The fleet launches 1 On-Demand Instance as specified, but needs to launch one more instance to fulfill the total target capacity. The purchasing option for the difference is calculated as `TotalTargetCapacity` – `OnDemandTargetCapacity` = `DefaultTargetCapacityType`, which results in the fleet launching 1 Spot Instance. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0e8c754449b27161c", "Version": "1" } } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 2, "OnDemandTargetCapacity": 1, "DefaultTargetCapacityType": "spot" } } ``` ## Example 4: Launch On-Demand Instances using multiple Capacity Reservations You can configure a fleet to use On-Demand Capacity Reservations first when launching On-Demand Instances by setting the usage strategy for Capacity Reservations to `use-capacity-reservations-first`. This example demonstrates how the fleet selects the Capacity Reservations to use when there are more Capacity Reservations than are needed to fulfil the target capacity. In this example, the fleet configuration is as follows: + Target capacity: 12 On-Demand Instances + Total unused Capacity Reservations: 15 (more than the fleet's target capacity of 12 On-Demand Instances) + Number of Capacity Reservation pools: 3 (`m5.large`, `m4.xlarge`, and `m4.2xlarge`) + Number of Capacity Reservations per pool: 5 + On-Demand allocation strategy: `lowest-price` (When there are multiple unused Capacity Reservations in multiple instance pools, the fleet determines the pools in which to launch the On-Demand Instances based on the On-Demand allocation strategy.) Note that you can also use the `prioritized` allocation strategy instead of the `lowest-price` allocation strategy. **Capacity Reservations** The account has the following 15 unused Capacity Reservations in 3 different pools. The number of Capacity Reservations in each pool is indicated by `AvailableInstanceCount`. ``` { "CapacityReservationId": "cr-111", "InstanceType": "m5.large", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount": 5, "InstanceMatchCriteria": "open", "State": "active" } { "CapacityReservationId": "cr-222", "InstanceType": "m4.xlarge", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount": 5, "InstanceMatchCriteria": "open", "State": "active" } { "CapacityReservationId": "cr-333", "InstanceType": "m4.2xlarge", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount":5, "InstanceMatchCriteria": "open", "State": "active" } ``` **Fleet configuration** The following fleet configuration shows only the pertinent configurations for this example. The total target capacity is 12, and the default target capacity type is `on-demand`. The On-Demand allocation strategy is `lowest-price`. The usage strategy for Capacity Reservations is `use-capacity-reservations-first`. In this example, the On-Demand Instance price is: + `m5.large` – \$10.096 per hour + `m4.xlarge` – \$10.20 per hour + `m4.2xlarge` – \$10.40 per hour **Note** The fleet type must be of type `instant`. Other fleet types do not support `use-capacity-reservations-first`. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-abc1234567example", "Version": "1" }, "Overrides": [ { "InstanceType": "m5.large", "AvailabilityZone": "us-east-1a", "WeightedCapacity": 1 }, { "InstanceType": "m4.xlarge", "AvailabilityZone": "us-east-1a", "WeightedCapacity": 1 }, { "InstanceType": "m4.2xlarge", "AvailabilityZone": "us-east-1a", "WeightedCapacity": 1 } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 12, "DefaultTargetCapacityType": "on-demand" }, "OnDemandOptions": { "AllocationStrategy": "lowest-price", "CapacityReservationOptions": { "UsageStrategy": "use-capacity-reservations-first" } }, "Type": "instant" } ``` After you create the `instant` fleet using the preceding configuration, the following 12 instances are launched to meet the target capacity: + 5 `m5.large` On-Demand Instances in `us-east-1a` – `m5.large` in `us-east-1a` is the lowest price, and there are 5 available unused `m5.large` Capacity Reservations + 5 `m4.xlarge` On-Demand Instances in us-east-1a – `m4.xlarge` in `us-east-1a` is the next lowest price, and there are 5 available unused `m4.xlarge` Capacity Reservations + 2 `m4.2xlarge` On-Demand Instances in us-east-1a – `m4.2xlarge` in `us-east-1a` is the third lowest price, and there are 5 available unused `m4.2xlarge` Capacity Reservations of which only 2 are needed to meet the target capacity After the fleet is launched, you can run [describe-capacity-reservations](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-capacity-reservations.html) to see how many unused Capacity Reservations are remaining. In this example, you should see the following response, which shows that all of the `m5.large` and `m4.xlarge` Capacity Reservations were used, with 3 `m4.2xlarge` Capacity Reservations remaining unused. ``` { "CapacityReservationId": "cr-111", "InstanceType": "m5.large", "AvailableInstanceCount": 0 } { "CapacityReservationId": "cr-222", "InstanceType": "m4.xlarge", "AvailableInstanceCount": 0 } { "CapacityReservationId": "cr-333", "InstanceType": "m4.2xlarge", "AvailableInstanceCount": 3 } ``` ## Example 5: Launch On-Demand Instances using Capacity Reservations when the total target capacity exceeds the number of unused Capacity Reservations You can configure a fleet to use On-Demand Capacity Reservations first when launching On-Demand Instances by setting the usage strategy for Capacity Reservations to `use-capacity-reservations-first`. This example demonstrates how the fleet selects the instance pools in which to launch On-Demand Instances when the total target capacity exceeds the number of available unused Capacity Reservations. In this example, the fleet configuration is as follows: + Target capacity: 16 On-Demand Instances + Total unused Capacity Reservations: 15 (less than the fleet's target capacity of 16 On-Demand Instances) + Number of Capacity Reservation pools: 3 (`m5.large`, `m4.xlarge`, and `m4.2xlarge`) + Number of Capacity Reservations per pool: 5 + On-Demand allocation strategy: `lowest-price` (When the number of unused Capacity Reservations is less than the On-Demand target capacity, the fleet determines the pools in which to launch the remaining On-Demand capacity based on the On-Demand allocation strategy.) Note that you can also use the `prioritized` allocation strategy instead of the `lowest-price` allocation strategy. **Capacity Reservations** The account has the following 15 unused Capacity Reservations in 3 different pools. The number of Capacity Reservations in each pool is indicated by `AvailableInstanceCount`. ``` { "CapacityReservationId": "cr-111", "InstanceType": "m5.large", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount": 5, "InstanceMatchCriteria": "open", "State": "active" } { "CapacityReservationId": "cr-222", "InstanceType": "m4.xlarge", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount": 5, "InstanceMatchCriteria": "open", "State": "active" } { "CapacityReservationId": "cr-333", "InstanceType": "m4.2xlarge", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount":5, "InstanceMatchCriteria": "open", "State": "active" } ``` **Fleet configuration** The following fleet configuration shows only the pertinent configurations for this example. The total target capacity is 16, and the default target capacity type is `on-demand`. The On-Demand allocation strategy is `lowest-price`. The usage strategy for Capacity Reservations is `use-capacity-reservations-first`. In this example, the On-Demand Instance price is: + m5.large – \$10.096 per hour + m4.xlarge – \$10.20 per hour + m4.2xlarge – \$10.40 per hour **Note** The fleet type must be `instant`. Other fleet types do not support `use-capacity-reservations-first`. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0e8c754449b27161c", "Version": "1" }, "Overrides": [ { "InstanceType": "m5.large", "AvailabilityZone": "us-east-1a", "WeightedCapacity": 1 }, { "InstanceType": "m4.xlarge", "AvailabilityZone": "us-east-1a", "WeightedCapacity": 1 }, { "InstanceType": "m4.2xlarge", "AvailabilityZone": "us-east-1a", "WeightedCapacity": 1 } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 16, "DefaultTargetCapacityType": "on-demand" }, "OnDemandOptions": { "AllocationStrategy": "lowest-price" "CapacityReservationOptions": { "UsageStrategy": "use-capacity-reservations-first" } }, "Type": "instant", } ``` After you create the `instant` fleet using the preceding configuration, the following 16 instances are launched to meet the target capacity: + 6 `m5.large` On-Demand Instances in `us-east-1a` – `m5.large` in `us-east-1a` is the lowest price, and there are 5 available unused `m5.large` Capacity Reservations. The Capacity Reservations are used first to launch 5 On-Demand Instances. After the remaining `m4.xlarge` and `m4.2xlarge` Capacity Reservations are used, to meet the target capacity an additional On-Demand Instance is launched according to the On-Demand allocation strategy, which is `lowest-price` in this example. + 5 `m4.xlarge` On-Demand Instances in `us-east-1a` – `m4.xlarge` in `us-east-1a` is the next lowest price, and there are 5 available unused `m4.xlarge` Capacity Reservations + 5 `m4.2xlarge` On-Demand Instances in `us-east-1a` – `m4.2xlarge` in `us-east-1a` is the third lowest price, and there are 5 available unused `m4.2xlarge` Capacity Reservations After the fleet is launched, you can run [describe-capacity-reservations](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-capacity-reservations.html) to see how many unused Capacity Reservations are remaining. In this example, you should see the following response, which shows that all of the Capacity Reservations in all of the pools were used. ``` { "CapacityReservationId": "cr-111", "InstanceType": "m5.large", "AvailableInstanceCount": 0 } { "CapacityReservationId": "cr-222", "InstanceType": "m4.xlarge", "AvailableInstanceCount": 0 } { "CapacityReservationId": "cr-333", "InstanceType": "m4.2xlarge", "AvailableInstanceCount": 0 } ``` ## Example 6: Launch On-Demand Instances using targeted Capacity Reservations You can configure a fleet to use `targeted` On-Demand Capacity Reservations first when launching On-Demand Instances by setting the usage strategy for Capacity Reservations to `use-capacity-reservations-first`. This example demonstrates how to launch On-Demand Instances into `targeted` Capacity Reservations, where the attributes of the Capacity Reservations are the same except for their Availability Zones (`us-east-1a` and `us-east-1b`). It also demonstrates how the fleet selects the instance pools in which to launch On-Demand Instances when the total target capacity exceeds the number of available unused Capacity Reservations. In this example, the fleet configuration is as follows: + Target capacity: 10 On-Demand Instances + Total unused `targeted` Capacity Reservations: 6 (less than the fleet's On-Demand target capacity of 10 On-Demand Instances) + Number of Capacity Reservation pools: 2 (`us-east-1a` and `us-east-1b`) + Number of Capacity Reservations per pool: 3 + On-Demand allocation strategy: `lowest-price` (When the number of unused Capacity Reservations is less than the On-Demand target capacity, the fleet determines the pools in which to launch the remaining On-Demand capacity based on the On-Demand allocation strategy.) Note that you can also use the `prioritized` allocation strategy instead of the `lowest-price` allocation strategy. For a walkthrough of the procedures that you must perform to accomplish this example, see [Tutorial: Configure EC2 Fleet to launch On-Demand Instances using targeted Capacity Reservations](ec2-fleet-launch-on-demand-instances-using-targeted-capacity-reservations-walkthrough.md). **Capacity Reservations** The account has the following 6 unused Capacity Reservations in 2 different pools. In this example, the pools differ by their Availability Zones. The number of Capacity Reservations in each pool is indicated by `AvailableInstanceCount`. ``` { "CapacityReservationId": "cr-111", "InstanceType": "c5.xlarge", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1a", "AvailableInstanceCount": 3, "InstanceMatchCriteria": "open", "State": "active" } { "CapacityReservationId": "cr-222", "InstanceType": "c5.xlarge", "InstancePlatform": "Linux/UNIX", "AvailabilityZone": "us-east-1b", "AvailableInstanceCount": 3, "InstanceMatchCriteria": "open", "State": "active" } ``` **Fleet configuration** The following fleet configuration shows only the pertinent configurations for this example. The total target capacity is 10, and the default target capacity type is `on-demand`. The On-Demand allocation strategy is `lowest-price`. The usage strategy for Capacity Reservations is `use-capacity-reservations-first`. In this example, the On-Demand Instance price for `c5.xlarge` in `us-east-1` is \$10.17 per hour. **Note** The fleet type must be `instant`. Other fleet types do not support `use-capacity-reservations-first`. ``` { "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "c5.xlarge", "AvailabilityZone": "us-east-1a" }, { "InstanceType": "c5.xlarge", "AvailabilityZone": "us-east-1b" } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 10, "DefaultTargetCapacityType": "on-demand" }, "OnDemandOptions": { "AllocationStrategy": "lowest-price", "CapacityReservationOptions": { "UsageStrategy": "use-capacity-reservations-first" } }, "Type": "instant" } ``` After you create the `instant` fleet using the preceding configuration, the following 10 instances are launched to meet the target capacity: + The Capacity Reservations are used first to launch 6 On-Demand Instances as follows: + 3 On-Demand Instances are launched into the 3 `c5.xlarge` `targeted` Capacity Reservations in `us-east-1a` + 3 On-Demand Instances are launched into the 3 `c5.xlarge` `targeted` Capacity Reservations in `us-east-1b` + To meet the target capacity, 4 additional On-Demand Instances are launched into regular On-Demand capacity according to the On-Demand allocation strategy, which is `lowest-price` in this example. However, because the pools are the same price (because price is per Region and not per Availability Zone), the fleet launches the remaining 4 On-Demand Instances into either of the pools. After the fleet is launched, you can run [describe-capacity-reservations](https://docs.aws.amazon.com/cli/latest/reference/ec2/describe-capacity-reservations.html) to see how many unused Capacity Reservations are remaining. In this example, you should see the following response, which shows that all of the Capacity Reservations in all of the pools were used. ``` { "CapacityReservationId": "cr-111", "InstanceType": "c5.xlarge", "AvailableInstanceCount": 0 } { "CapacityReservationId": "cr-222", "InstanceType": "c5.xlarge", "AvailableInstanceCount": 0 } ``` ## Example 7: Configure Capacity Rebalancing to launch replacement Spot Instances The following example configures the EC2 Fleet to launch a replacement Spot Instance when Amazon EC2 emits a rebalance recommendation for a Spot Instance in the fleet. To configure the automatic replacement of Spot Instances, for `ReplacementStrategy`, specify `launch-before-terminate`. To configure the time delay from when the new replacement Spot Instances are launched to when the old Spot Instances are automatically deleted, for `termination-delay`, specify a value in seconds. For more information, see [Configuration options](ec2-fleet-capacity-rebalance.md#ec2-fleet-capacity-rebalance-config-options). **Note** We recommend using `launch-before-terminate` only if you can predict how long your instance shutdown procedures will take to complete so that the old instances are only terminated after these procedures are completed. You are charged for all instances while they are running. The effectiveness of the Capacity Rebalancing strategy depends on the number of Spot capacity pools specified in the EC2 Fleet request. We recommend that you configure the fleet with a diversified set of instance types and Availability Zones, and for `AllocationStrategy`, specify `capacity-optimized`. For more information about what you should consider when configuring an EC2 Fleet for Capacity Rebalancing, see [Use Capacity Rebalancing in EC2 Fleet and Spot Fleet to replace at-risk Spot Instances](ec2-fleet-capacity-rebalance.md). ``` { "ExcessCapacityTerminationPolicy": "termination", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "LaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceType": "c3.large", "WeightedCapacity": 1, "Placement": { "AvailabilityZone": "us-east-1a" } }, { "InstanceType": "c4.large", "WeightedCapacity": 1, "Placement": { "AvailabilityZone": "us-east-1a" } }, { "InstanceType": "c5.large", "WeightedCapacity": 1, "Placement": { "AvailabilityZone": "us-east-1a" } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 5, "DefaultTargetCapacityType": "spot" }, "SpotOptions": { "AllocationStrategy": "capacity-optimized", "MaintenanceStrategies": { "CapacityRebalance": { "ReplacementStrategy": "launch-before-terminate", "TerminationDelay": "720" } } } } ``` ## Example 8: Launch Spot Instances in a capacity-optimized fleet The following example demonstrates how to configure an EC2 Fleet with a Spot allocation strategy that optimizes for capacity. To optimize for capacity, you must set `AllocationStrategy` to `capacity-optimized`. In the following example, the three launch specifications specify three Spot capacity pools. The target capacity is 50 Spot Instances. The EC2 Fleet attempts to launch 50 Spot Instances into the Spot capacity pool with optimal capacity for the number of instances that are launching. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized", }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "r4.2xlarge", "Placement": { "AvailabilityZone": "us-west-2a" }, }, { "InstanceType": "m4.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" }, }, { "InstanceType": "c5.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 50, "DefaultTargetCapacityType": "spot" } } ``` ## Example 9: Launch Spot Instances in a capacity-optimized fleet with priorities The following example demonstrates how to configure an EC2 Fleet with a Spot allocation strategy that optimizes for capacity while using priority on a best-effort basis. When using the `capacity-optimized-prioritized` allocation strategy, you can use the `Priority` parameter to specify the priorities of the Spot capacity pools, where the lower the number the higher priority. You can also set the same priority for several Spot capacity pools if you favor them equally. If you do not set a priority for a pool, the pool will be considered last in terms of priority. To prioritize Spot capacity pools, you must set `AllocationStrategy` to `capacity-optimized-prioritized`. EC2 Fleet will optimize for capacity first, but will honor the priorities on a best-effort basis (for example, if honoring the priorities will not significantly affect EC2 Fleet's ability to provision optimal capacity). This is a good option for workloads where the possibility of disruption must be minimized and the preference for certain instance types matters. In the following example, the three launch specifications specify three Spot capacity pools. Each pool is prioritized, where the lower the number the higher priority. The target capacity is 50 Spot Instances. The EC2 Fleet attempts to launch 50 Spot Instances into the Spot capacity pool with the highest priority on a best-effort basis, but optimizes for capacity first. ``` { "SpotOptions": { "AllocationStrategy": "capacity-optimized-prioritized" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "r4.2xlarge", "Priority": 1, "Placement": { "AvailabilityZone": "us-west-2a" }, }, { "InstanceType": "m4.2xlarge", "Priority": 2, "Placement": { "AvailabilityZone": "us-west-2b" }, }, { "InstanceType": "c5.2xlarge", "Priority": 3, "Placement": { "AvailabilityZone": "us-west-2b" } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 50, "DefaultTargetCapacityType": "spot" } ``` ## Example 10: Launch Spot Instances in a price-capacity-optimized fleet The following example demonstrates how to configure an EC2 Fleet with a Spot allocation strategy that optimizes for both capacity and lowest price. To optimize for capacity while taking price into consideration, you must set the Spot `AllocationStrategy` to `price-capacity-optimized`. In the following example, the three launch specifications specify three Spot capacity pools. The target capacity is 50 Spot Instances. The EC2 Fleet attempts to launch 50 Spot Instances into the Spot capacity pool with optimal capacity for the number of instances that are launching while also choosing the pool that is the lowest priced. ``` { "SpotOptions": { "AllocationStrategy": "price-capacity-optimized", "MinTargetCapacity": 2, "SingleInstanceType": true }, "OnDemandOptions": { "AllocationStrategy": "lowest-price" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "r4.2xlarge", "Placement": { "AvailabilityZone": "us-west-2a" }, }, { "InstanceType": "m4.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" }, }, { "InstanceType": "c5.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } } ] } ], "TargetCapacitySpecification": { "TotalTargetCapacity": 50, "OnDemandTargetCapacity":0, "SpotTargetCapacity":50, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` ## Example 11: Configure attribute-based instance type selection The following example demonstrates how to configure an EC2 Fleet to use attribute-based instance type selection for identifying instance types. To specify the required instance attributes, you specify the attributes in the `InstanceRequirements` structure. In the following example, two instance attributes are specified: + `VCpuCount` – A minimum of 2 vCPUs is specified. Because no maximum is specified, there is no maximum limit. + `MemoryMiB` – A minimum of 4 MiB of memory is specified. Because no maximum is specified, there is no maximum limit. Any instance types that have 2 or more vCPUs and 4 MiB or more of memory will be identified. However, price protection and the allocation strategy might exclude some instance types when [EC2 Fleet provisions the fleet](ec2-fleet-attribute-based-instance-type-selection.md#how-ef-uses-abs). For a list and descriptions of all the possible attributes that you can specify, see [InstanceRequirements](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_InstanceRequirements.html) in the *Amazon EC2 API Reference*. ``` { "SpotOptions": { "AllocationStrategy": "price-capacity-optimized" }, "LaunchTemplateConfigs": [{ "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [{ "InstanceRequirements": { "VCpuCount": { "Min": 2 }, "MemoryMiB": { "Min": 4 } } }] }], "TargetCapacitySpecification": { "TotalTargetCapacity": 20, "DefaultTargetCapacityType": "spot" }, "Type": "instant" } ``` # Example CLI configurations Spot Fleet You can define your Spot Fleet configuration in a JSON file, and then reference that file using the [request-spot-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/request-spot-fleet.html) AWS CLI command to create your fleet, as follows: ``` aws ec2 request-spot-fleet --spot-fleet-request-config file://file_name.json ``` The following examples illustrate launch configurations for various Spot Fleet use cases. For more information about the configuration parameters, see [request-spot-fleet](https://docs.aws.amazon.com/cli/latest/reference/ec2/request-spot-fleet.html). For more information about creating a Spot Fleet, see [Create a Spot Fleet](create-spot-fleet.md). **Note** For Spot Fleet, you can't specify a network interface ID in a launch template or launch specification. Make sure you omit the `NetworkInterfaceID` parameter in your launch template or launch specification. **Topics** + [ ## Example 1: Launch Spot Instances using the lowest-priced Availability Zone in the Region ](#fleet-config1) + [ ## Example 2: Launch Spot Instances using the lowest-priced Availability Zone from a list ](#fleet-config2) + [ ## Example 3: Launch Spot Instances using the lowest-priced instance type from a list ](#fleet-config3) + [ ## Example 4. Override the price for the request ](#fleet-config4) + [ ## Example 5: Launch a Spot Fleet using the diversified allocation strategy ](#fleet-config5) + [ ## Example 6: Launch a Spot Fleet using instance weighting ](#fleet-config6) + [ ## Example 7: Launch a Spot Fleet using On-Demand capacity ](#fleet-config7) + [ ## Example 8: Configure Capacity Rebalancing to launch replacement Spot Instances ](#fleet-config8) + [ ## Example 9: Launch Spot Instances in a capacity-optimized fleet ](#fleet-config9) + [ ## Example 10: Launch Spot Instances in a capacity-optimized fleet with priorities ](#fleet-config10) + [ ## Example 11: Launch Spot Instances in a priceCapacityOptimized fleet ](#fleet-config11) + [ ## Example 12: Configure attribute-based instance type selection ](#fleet-config12) ## Example 1: Launch Spot Instances using the lowest-priced Availability Zone in the Region The following example specifies a single launch specification without an Availability Zone or subnet. The Spot Fleet launches the instances in the lowest-priced Availability Zone that has a default subnet. The price you pay does not exceed the On-Demand price. ``` { "TargetCapacity": 20, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "KeyName": "my-key-pair", "SecurityGroups": [ { "GroupId": "sg-1a2b3c4d" } ], "InstanceType": "m3.medium", "IamInstanceProfile": { "Arn": "arn:aws:iam::123456789012:instance-profile/my-iam-role" } } ] } ``` ## Example 2: Launch Spot Instances using the lowest-priced Availability Zone from a list The following examples specify two launch specifications with different Availability Zones or subnets, but the same instance type and AMI. **Availability Zones** The Spot Fleet launches the instances in the default subnet of the lowest-priced Availability Zone that you specified. ``` { "TargetCapacity": 20, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "KeyName": "my-key-pair", "SecurityGroups": [ { "GroupId": "sg-1a2b3c4d" } ], "InstanceType": "m3.medium", "Placement": { "AvailabilityZone": "us-west-2a, us-west-2b" }, "IamInstanceProfile": { "Arn": "arn:aws:iam::123456789012:instance-profile/my-iam-role" } } ] } ``` **Subnets** You can specify default subnets or nondefault subnets, and the nondefault subnets can be from a default VPC or a nondefault VPC. The Spot service launches the instances in whichever subnet is in the lowest-priced Availability Zone. You can't specify different subnets from the same Availability Zone in a Spot Fleet request. ``` { "TargetCapacity": 20, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "KeyName": "my-key-pair", "SecurityGroups": [ { "GroupId": "sg-1a2b3c4d" } ], "InstanceType": "m3.medium", "SubnetId": "subnet-a61dafcf, subnet-65ea5f08", "IamInstanceProfile": { "Arn": "arn:aws:iam::123456789012:instance-profile/my-iam-role" } } ] } ``` If the instances are launched in a default VPC, they receive a public IPv4 address by default. If the instances are launched in a nondefault VPC, they do not receive a public IPv4 address by default. Use a network interface in the launch specification to assign a public IPv4 address to instances launched in a nondefault VPC. When you specify a network interface, you must include the subnet ID and security group ID using the network interface. ``` ... { "ImageId": "ami-1a2b3c4d", "KeyName": "my-key-pair", "InstanceType": "m3.medium", "NetworkInterfaces": [ { "DeviceIndex": 0, "SubnetId": "subnet-1a2b3c4d", "Groups": [ "sg-1a2b3c4d" ], "AssociatePublicIpAddress": true } ], "IamInstanceProfile": { "Arn": "arn:aws:iam::123456789012:instance-profile/my-iam-role" } } ... ``` ## Example 3: Launch Spot Instances using the lowest-priced instance type from a list The following examples specify two launch configurations with different instance types, but the same AMI and Availability Zone or subnet. The Spot Fleet launches the instances using the specified instance type with the lowest price. **Availability Zone** ``` { "TargetCapacity": 20, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "SecurityGroups": [ { "GroupId": "sg-1a2b3c4d" } ], "InstanceType": "c5.4xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } }, { "ImageId": "ami-1a2b3c4d", "SecurityGroups": [ { "GroupId": "sg-1a2b3c4d" } ], "InstanceType": "r3.8xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } } ] } ``` **Subnet** ``` { "TargetCapacity": 20, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "SecurityGroups": [ { "GroupId": "sg-1a2b3c4d" } ], "InstanceType": "c5.4xlarge", "SubnetId": "subnet-1a2b3c4d" }, { "ImageId": "ami-1a2b3c4d", "SecurityGroups": [ { "GroupId": "sg-1a2b3c4d" } ], "InstanceType": "r3.8xlarge", "SubnetId": "subnet-1a2b3c4d" } ] } ``` ## Example 4. Override the price for the request We recommended that you use the default maximum price, which is the On-Demand price. If you prefer, you can specify a maximum price for the fleet request and maximum prices for individual launch specifications. The following examples specify a maximum price for the fleet request and maximum prices for two of the three launch specifications. The maximum price for the fleet request is used for any launch specification that does not specify a maximum price. The Spot Fleet launches the instances using the instance type with the lowest price. **Availability Zone** ``` { "SpotPrice": "1.00", "TargetCapacity": 30, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "InstanceType": "c3.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" }, "SpotPrice": "0.10" }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "c3.4xlarge", "Placement": { "AvailabilityZone": "us-west-2b" }, "SpotPrice": "0.20" }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "c3.8xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } } ] } ``` **Subnet** ``` { "SpotPrice": "1.00", "TargetCapacity": 30, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "InstanceType": "c3.2xlarge", "SubnetId": "subnet-1a2b3c4d", "SpotPrice": "0.10" }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "c3.4xlarge", "SubnetId": "subnet-1a2b3c4d", "SpotPrice": "0.20" }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "c3.8xlarge", "SubnetId": "subnet-1a2b3c4d" } ] } ``` ## Example 5: Launch a Spot Fleet using the diversified allocation strategy The following example uses the `diversified` allocation strategy. The launch specifications have different instance types but the same AMI and Availability Zone or subnet. The Spot Fleet distributes the 30 instances across the three launch specifications, such that there are 10 instances of each type. For more information, see [Use allocation strategies to determine how EC2 Fleet or Spot Fleet fulfills Spot and On-Demand capacity](ec2-fleet-allocation-strategy.md). **Availability Zone** ``` { "SpotPrice": "0.70", "TargetCapacity": 30, "AllocationStrategy": "diversified", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "InstanceType": "c4.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "m3.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "r3.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } } ] } ``` **Subnet** ``` { "SpotPrice": "0.70", "TargetCapacity": 30, "AllocationStrategy": "diversified", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "InstanceType": "c4.2xlarge", "SubnetId": "subnet-1a2b3c4d" }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "m3.2xlarge", "SubnetId": "subnet-1a2b3c4d" }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "r3.2xlarge", "SubnetId": "subnet-1a2b3c4d" } ] } ``` A best practice to increase the chance that a spot request can be fulfilled by EC2 capacity in the event of an outage in one of the Availability Zones is to diversify across zones. For this scenario, include each Availability Zone available to you in the launch specification. And, instead of using the same subnet each time, use three unique subnets (each mapping to a different zone). **Availability Zone** ``` { "SpotPrice": "0.70", "TargetCapacity": 30, "AllocationStrategy": "diversified", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "InstanceType": "c4.2xlarge", "Placement": { "AvailabilityZone": "us-west-2a" } }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "m3.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" } }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "r3.2xlarge", "Placement": { "AvailabilityZone": "us-west-2c" } } ] } ``` **Subnet** ``` { "SpotPrice": "0.70", "TargetCapacity": 30, "AllocationStrategy": "diversified", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "InstanceType": "c4.2xlarge", "SubnetId": "subnet-1a2b3c4d" }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "m3.2xlarge", "SubnetId": "subnet-2a2b3c4d" }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "r3.2xlarge", "SubnetId": "subnet-3a2b3c4d" } ] } ``` ## Example 6: Launch a Spot Fleet using instance weighting The following examples use instance weighting, which means that the price is per unit hour instead of per instance hour. Each launch configuration lists a different instance type and a different weight. The Spot Fleet selects the instance type with the lowest price per unit hour. The Spot Fleet calculates the number of Spot Instances to launch by dividing the target capacity by the instance weight. If the result isn't an integer, the Spot Fleet rounds it up to the next integer, so that the size of your fleet is not below its target capacity. If the `r3.2xlarge` request is successful, Spot provisions 4 of these instances. Divide 20 by 6 for a total of 3.33 instances, then round up to 4 instances. If the `c3.xlarge` request is successful, Spot provisions 7 of these instances. Divide 20 by 3 for a total of 6.66 instances, then round up to 7 instances. For more information, see [Use instance weighting to manage cost and performance of your EC2 Fleet or Spot Fleet](ec2-fleet-instance-weighting.md). **Availability Zone** ``` { "SpotPrice": "0.70", "TargetCapacity": 20, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "InstanceType": "r3.2xlarge", "Placement": { "AvailabilityZone": "us-west-2b" }, "WeightedCapacity": 6 }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "c3.xlarge", "Placement": { "AvailabilityZone": "us-west-2b" }, "WeightedCapacity": 3 } ] } ``` **Subnet** ``` { "SpotPrice": "0.70", "TargetCapacity": 20, "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchSpecifications": [ { "ImageId": "ami-1a2b3c4d", "InstanceType": "r3.2xlarge", "SubnetId": "subnet-1a2b3c4d", "WeightedCapacity": 6 }, { "ImageId": "ami-1a2b3c4d", "InstanceType": "c3.xlarge", "SubnetId": "subnet-1a2b3c4d", "WeightedCapacity": 3 } ] } ``` ## Example 7: Launch a Spot Fleet using On-Demand capacity To ensure that you always have instance capacity, you can include a request for On-Demand capacity in your Spot Fleet request. If there is capacity, the On-Demand request is always fulfilled. The balance of the target capacity is fulfilled as Spot if there is capacity and availability. The following example specifies the desired target capacity as 10, of which 5 must be On-Demand capacity. Spot capacity is not specified; it is implied in the balance of the target capacity minus the On-Demand capacity. Amazon EC2 launches 5 capacity units as On-Demand, and 5 capacity units (10-5=5) as Spot if there is available Amazon EC2 capacity and availability. ``` { "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "AllocationStrategy": "lowestPrice", "TargetCapacity": 10, "SpotPrice": null, "ValidFrom": "2018-04-04T15:58:13Z", "ValidUntil": "2019-04-04T15:58:13Z", "TerminateInstancesWithExpiration": true, "LaunchSpecifications": [], "Type": "maintain", "OnDemandTargetCapacity": 5, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0dbb04d4a6cca5ad1", "Version": "2" }, "Overrides": [ { "InstanceType": "t2.medium", "WeightedCapacity": 1, "SubnetId": "subnet-d0dc51fb" } ] } ] } ``` ## Example 8: Configure Capacity Rebalancing to launch replacement Spot Instances The following example configures the Spot Fleet to launch a replacement Spot Instance when Amazon EC2 emits a rebalance recommendation for a Spot Instance in the fleet. To configure the automatic replacement of Spot Instances, for `ReplacementStrategy`, specify `launch-before-terminate`. To configure the time delay from the launch of the new replacement Spot Instances to the automatic deletion of the old Spot Instances, for `termination-delay`, specify a value in seconds. For more information, see [Configuration options](ec2-fleet-capacity-rebalance.md#ec2-fleet-capacity-rebalance-config-options). **Note** We recommend using `launch-before-terminate` only if you can predict how long your instance shutdown procedures will take to complete. This ensures that the old instances are terminated only after the shutdown procedures are completed. You are charged for all instances while they are running. The effectiveness of the Capacity Rebalancing strategy depends on the number of Spot capacity pools specified in the Spot Fleet request. We recommend that you configure the fleet with a diversified set of instance types and Availability Zones, and for `AllocationStrategy`, specify `capacityOptimized`. For more information about what you should consider when configuring a Spot Fleet for Capacity Rebalancing, see [Use Capacity Rebalancing in EC2 Fleet and Spot Fleet to replace at-risk Spot Instances](ec2-fleet-capacity-rebalance.md). ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "capacityOptimized", "IamFleetRole": "arn:aws:iam::123456789012:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "LaunchTemplate", "Version": "1" }, "Overrides": [ { "InstanceType": "c3.large", "WeightedCapacity": 1, "Placement": { "AvailabilityZone": "us-east-1a" } }, { "InstanceType": "c4.large", "WeightedCapacity": 1, "Placement": { "AvailabilityZone": "us-east-1a" } }, { "InstanceType": "c5.large", "WeightedCapacity": 1, "Placement": { "AvailabilityZone": "us-east-1a" } } ] } ], "TargetCapacity": 5, "SpotMaintenanceStrategies": { "CapacityRebalance": { "ReplacementStrategy": "launch-before-terminate", "TerminationDelay": "720" } } } } ``` ## Example 9: Launch Spot Instances in a capacity-optimized fleet The following example demonstrates how to configure a Spot Fleet with a Spot allocation strategy that optimizes for capacity. To optimize for capacity, you must set `AllocationStrategy` to `capacityOptimized`. In the following example, the three launch specifications specify three Spot capacity pools. The target capacity is 50 Spot Instances. The Spot Fleet attempts to launch 50 Spot Instances into the Spot capacity pool with optimal capacity for the number of instances that are launching. ``` { "TargetCapacity": "50", "SpotFleetRequestConfig": { "AllocationStrategy": "capacityOptimized", }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "r4.2xlarge", "AvailabilityZone": "us-west-2a" }, { "InstanceType": "m4.2xlarge", "AvailabilityZone": "us-west-2b" }, { "InstanceType": "c5.2xlarge", "AvailabilityZone": "us-west-2b" } ] } ] } ``` ## Example 10: Launch Spot Instances in a capacity-optimized fleet with priorities The following example demonstrates how to configure a Spot Fleet with a Spot allocation strategy that optimizes for capacity while using priority on a best-effort basis. When using the `capacityOptimizedPrioritized` allocation strategy, you can use the `Priority` parameter to specify the priorities of the Spot capacity pools, where the lower the number the higher priority. You can also set the same priority for several Spot capacity pools if you favor them equally. If you do not set a priority for a pool, the pool will be considered last in terms of priority. To prioritize Spot capacity pools, you must set `AllocationStrategy` to `capacityOptimizedPrioritized`. Spot Fleet will optimize for capacity first, but will honor the priorities on a best-effort basis (for example, if honoring the priorities will not significantly affect Spot Fleet's ability to provision optimal capacity). This is a good option for workloads where the possibility of disruption must be minimized and the preference for certain instance types matters. In the following example, the three launch specifications specify three Spot capacity pools. Each pool is prioritized, where the lower the number the higher priority. The target capacity is 50 Spot Instances. The Spot Fleet attempts to launch 50 Spot Instances into the Spot capacity pool with the highest priority on a best-effort basis, but optimizes for capacity first. ``` { "TargetCapacity": "50", "SpotFleetRequestConfig": { "AllocationStrategy": "capacityOptimizedPrioritized" }, "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [ { "InstanceType": "r4.2xlarge", "Priority": 1, "AvailabilityZone": "us-west-2a" }, { "InstanceType": "m4.2xlarge", "Priority": 2, "AvailabilityZone": "us-west-2b" }, { "InstanceType": "c5.2xlarge", "Priority": 3, "AvailabilityZone": "us-west-2b" } ] } ] } ``` ## Example 11: Launch Spot Instances in a priceCapacityOptimized fleet The following example demonstrates how to configure a Spot Fleet with a Spot allocation strategy that optimizes for both capacity and lowest price. To optimize for capacity while taking price into consideration, you must set the Spot `AllocationStrategy` to `priceCapacityOptimized`. In the following example, the three launch specifications specify three Spot capacity pools. The target capacity is 50 Spot Instances. The Spot Fleet attempts to launch 50 Spot Instances into the Spot capacity pool with optimal capacity for the number of instances that are launching while also choosing the pool that is the lowest priced. ``` { "SpotFleetRequestConfig": { "AllocationStrategy": "priceCapacityOptimized", "OnDemandAllocationStrategy": "lowestPrice", "ExcessCapacityTerminationPolicy": "default", "IamFleetRole": "arn:aws:iam::111111111111:role/aws-ec2-spot-fleet-tagging-role", "LaunchTemplateConfigs": [ { "LaunchTemplateSpecification": { "LaunchTemplateId": "lt-0123456789example", "Version": "1" }, "Overrides": [ { "InstanceType": "r4.2xlarge", "AvailabilityZone": "us-west-2a" }, { "InstanceType": "m4.2xlarge", "AvailabilityZone": "us-west-2b" }, { "InstanceType": "c5.2xlarge", "AvailabilityZone": "us-west-2b" } ] } ], "TargetCapacity": 50, "Type": "request" } } ``` ## Example 12: Configure attribute-based instance type selection The following example demonstrates how to configure a Spot Fleet to use attribute-based instance type selection for identifying instance types. To specify the required instance attributes, you specify the attributes in the `InstanceRequirements` structure. In the following example, two instance attributes are specified: + `VCpuCount` – A minimum of 2 vCPUs is specified. Because no maximum is specified, there is no maximum limit. + `MemoryMiB` – A minimum of 4 MiB of memory is specified. Because no maximum is specified, there is no maximum limit. Any instance types that have 2 or more vCPUs and 4 MiB or more of memory will be identified. However, price protection and the allocation strategy might exclude some instance types when [Spot Fleet provisions the fleet](ec2-fleet-attribute-based-instance-type-selection.md#how-ef-uses-abs). For a list and descriptions of all the possible attributes that you can specify, see [InstanceRequirements](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_InstanceRequirements.html) in the *Amazon EC2 API Reference*. ``` { "AllocationStrategy": "priceCapacityOptimized", "TargetCapacity": 20, "Type": "request", "LaunchTemplateConfigs": [{ "LaunchTemplateSpecification": { "LaunchTemplateName": "my-launch-template", "Version": "1" }, "Overrides": [{ "InstanceRequirements": { "VCpuCount": { "Min": 2 }, "MemoryMiB": { "Min": 4 } } }] }] } ``` # Quotas for EC2 Fleet and Spot Fleet Your AWS account has default quotas, formerly referred to as limits, for each AWS service. Unless otherwise noted, each quota is Region-specific. The usual Amazon EC2 quotas apply to instances launched by an EC2 Fleet or a Spot Fleet, such as [Spot Instance limits](using-spot-limits.md) and [volume limits](volume_limits.md). In addition, your AWS account has the following quotas related to EC2 Fleet and Spot Fleet: **** | Quota description | Quota | | --- | --- | | The number of EC2 Fleets and Spot Fleets per Region of type maintain and request in the active, deleted\$1running, and cancelled\$1running states | 1,000 ¹ ² ³ | | The number of EC2 Fleets of type instant | Unlimited | | The number of Spot capacity pools (unique combination of instance type and subnet) for EC2 Fleets and Spot Fleets of type maintain and request | 300 ¹ | | The number of Spot capacity pools (unique combination of instance type and subnet) for EC2 Fleets of type instant | Unlimited | | The size of the user data in a launch specification | 16 KB ² | | The target capacity per EC2 Fleet or Spot Fleet | 10,000 | | The target capacity across all EC2 Fleets and Spot Fleets in a Region | 100,000 ¹ | | An EC2 Fleet request or a Spot Fleet request can't span Regions. | | | An EC2 Fleet request or a Spot Fleet request can't span different subnets from the same Availability Zone. | | ¹ These quotas apply to both your EC2 Fleets and your Spot Fleets. ² These are hard quotas. You can't request an increase for these quotas. ³ After you delete an EC2 Fleet or cancel a Spot Fleet request, and if you specified that the fleet should *not* terminate its Spot Instances when you deleted or canceled the request, the fleet request enters the `deleted_running` (EC2 Fleet) or `cancelled_running` (Spot Fleet) state and the instances continue to run until they are interrupted or you terminate them manually. If you terminate the instances, the fleet request enters the `deleted_terminating` (EC2 Fleet) or `cancelled_terminating` ( Spot Fleet) state and does not count towards this quota. For more information, see [Delete an EC2 Fleet request and the instances in the fleet](delete-fleet.md) and [Cancel (delete) a Spot Fleet request](cancel-spot-fleet.md). ## Request a quota increase for target capacity If you need more than the default quota for target capacity, you can request a quota increase. **To request a quota increase for target capacity** 1. Open the Support Center [Create case](https://console.aws.amazon.com/support/home#/case/create?issueType=service-limit-increase&limitType=service-code-ec2-fleet) form. 1. Choose **Service limit increase**. 1. For **Limit type**, choose **EC2 Fleet**. 1. For **Region**, choose the AWS Region in which to request the quota increase. 1. For **Limit**, choose **Target Fleet Capacity per Fleet (in units)** or **Target Fleet Capacity per Region (in units)**, depending on which quota you want to increase. 1. For **New limit value**, enter the new quota value. 1. To request an increase for another quota, choose **Add another request**, and repeat Steps 4–6. 1. For **Use case description**, enter your reason for requesting a quota increase. 1. Under **Contact options**, specify your preferred contact language and contact method. 1. Choose **Submit**.