Reserve capacity in specific Availability Zones with Capacity Reservations - Amazon EC2 Auto Scaling
Step 1: Create the Capacity ReservationsStep 2: Create a Capacity Reservation groupStep 3: Create a launch templateStep 4: Create an Auto Scaling groupRelated resources

Reserve capacity in specific Availability Zones with Capacity Reservations

Amazon EC2 On-Demand Capacity Reservations help you reserve compute capacity in specific Availability Zones. To start using Capacity Reservations, you create a Capacity Reservation in a specific Availability Zone. Then, you can launch instances into the reserved capacity, view its capacity utilization in real time, and increase or decrease its capacity as needed.

Capacity Reservations are configured as either open or targeted. If the Capacity Reservation is open, all new and existing instances that have matching attributes automatically run in the capacity of the Capacity Reservation. If the Capacity Reservation is targeted, instances must specifically target it to run in the reserved capacity.

This topic shows how to create an Auto Scaling group that launches On-Demand Instances into targeted Capacity Reservations. This gives you more control over when to use specific Capacity Reservations.

The basic steps are:

  1. Create Capacity Reservations in multiple Availability Zones that have the same instance type, platform, and instance count.

  2. Group Capacity Reservations using AWS Resource Groups.

  3. Create an Auto Scaling group with a launch template that targets the resource group, using the same Availability Zones as the Capacity Reservations.

Step 1: Create the Capacity Reservations

The first step is to create a Capacity Reservation in each Availability Zone where your Auto Scaling group will be deployed.

Note

You can only create targeted reservations when you first create the Capacity Reservations.

Console
To create your Capacity Reservations

  1. Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/.

  2. Choose Capacity Reservations, and then choose Create Capacity Reservation.

  3. On the Create a Capacity Reservation page, pay attention to following settings in the Instance details section. The instance type, platform, and Availability Zone of the instances that you launch must match the instance type, platform, and Availability Zone that you specify here or the Capacity Reservation is not applied.

    1. For Instance type, choose the type of instance to launch into the reserved capacity.

    2. For Platform, choose the operating system for your instances.

    3. For Availability Zone, choose the first Availability Zone that you want to reserve capacity in.

    4. For Total capacity, choose the number of instances you need. Calculate the total number of instances you need for your Auto Scaling group divided by the number of Availability Zones you plan to use.

  4. Under Capacity Reservation details, for Capacity Reservation ends, choose one of the following options:

    • At specific time – Cancel the Capacity Reservation automatically at the specified date and time.

    • Manually – Reserve the capacity until you explicitly cancel it.

  5. For Instance eligibility, choose Targeted: Only instances that target the Capacity Reservation.

  6. (Optional) For Tags, specify any tags to associate with the Capacity Reservation.

  7. Choose Create.

  8. Note the ID of the newly created Capacity Reservation. You need it to set up the Capacity Reservation group.

Repeat this procedure for each Availability Zone you want to enable for your Auto Scaling group, changing only the value of the Availability Zone option.

AWS CLI
To create your Capacity Reservations

Use the following create-capacity-reservation command to create the Capacity Reservations. Replace the sample values for --availability-zone, --instance-type, --instance-platform, and --instance-count. 

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

{
    "CapacityReservation": {
        "CapacityReservationId": "cr-1234567890abcdef1",
        "OwnerId": "123456789012",
        "CapacityReservationArn": "arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-1234567890abcdef1",
        "InstanceType": "c5.xlarge",
        "InstancePlatform": "Linux/UNIX",
        "AvailabilityZone": "us-east-1a",
        "Tenancy": "default",
        "TotalInstanceCount": 3,
        "AvailableInstanceCount": 3,
        "EbsOptimized": false,
        "EphemeralStorage": false,
        "State": "active",
        "StartDate": "2023-07-26T21:36:14+00:00",
        "EndDateType": "unlimited",
        "InstanceMatchCriteria": "targeted",
        "CreateDate": "2023-07-26T21:36:14+00:00"
    }
}

Note the ID of the newly created Capacity Reservation. You need it to set up the Capacity Reservation group.

Repeat this command for each Availability Zone you want to enable for your Auto Scaling group, changing only the value of the --availability-zone option.

Step 2: Create a Capacity Reservation group

When you finish creating the Capacity Reservations, you can group them together using the AWS Resource Groups service. AWS Resource Groups supports several different types of groups for different uses. Amazon EC2 uses a special-purpose group, known as a service-linked resource group, to target a group of Capacity Reservations. To interact with this service-linked resource group, you can use the AWS CLI or an SDK but not the console. For more information about service-linked resource groups, see Service configurations for resource groups in the AWS Resource Groups User Guide.

To create a Capacity Reservation group using the AWS CLI

Use the create-group command to create a resource group that can contain only Capacity Reservations. In this example, the resource group is named my-cr-group.

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"]}]}'

The following is an example response.

{
    "Group": {
        "GroupArn": "arn:aws:resource-groups:us-east-1:123456789012:group/my-cr-group",
        "Name": "my-cr-group"
    },
    "GroupConfiguration": {
        "Configuration": [
            {
                "Type": "AWS::EC2::CapacityReservationPool"
            },
            {
                "Type": "AWS::ResourceGroups::Generic",
                "Parameters": [
                    {
                        "Name": "allowed-resource-types",
                        "Values": [
                            "AWS::EC2::CapacityReservation"
                        ]
                    }
                ]
            }
        ],
        "Status": "UPDATE_COMPLETE"
    }
}

Note the ARN of the resource group. You need it to set up the launch template for your Auto Scaling group.

To associate your Capacity Reservations to the newly created group using the AWS CLI

Use the following group-resources command to associate the Capacity Reservations to the newly created Capacity Reservation group. For the --resource-arns option, specify the Capacity Reservations using their ARNs. Construct the ARNs using the relevant Region, your account ID, and the reservation IDs you noted earlier. In this example, the reservations with IDs cr-1234567890abcdef1 and cr-54321abcdef567890 will be grouped together in the group named my-cr-group. 

aws resource-groups group-resources \
    --group my-cr-group \
    --resource-arns \
      arn:aws:ec2:region:account-id:capacity-reservation/cr-1234567890abcdef1 \
      arn:aws:ec2:region:account-id:capacity-reservation/cr-54321abcdef567890

The following is an example response.

{
   "Succeeded": [ 
   "arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-1234567890abcdef1", 
   "arn:aws:ec2:us-east-1:123456789012:capacity-reservation/cr-54321abcdef567890" 
   ],
   "Failed": [],
   "Pending": []
}

For information about modifying or deleting the resource group, see the AWS Resource Groups API Reference.

Step 3: Create a launch template

Console
To create a launch template

  1. Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/.

  2. On the navigation pane, under Instances, choose Launch Templates.

  3. Choose Create launch template. Enter a name and provide a description for the initial version of the launch template.

  4. Under Auto Scaling guidance, select the check box.

  5. Create the launch template. Choose an AMI and instance type that matches the Capacity Reservations you are planning to use, and optionally, a key pair, one or more security groups, and any additional EBS volumes or instance store volumes for your instances.

  6. Expand Advanced details, and do the following:

    1. For Capacity Reservation, choose Target by group.

    2. For Capacity reservation - Target by group, choose the Capacity Reservations group that you created in the previous section, and then choose Save.

  7. Choose Create launch template.

  8. On the confirmation page, choose Create Auto Scaling group.

AWS CLI
To create a launch template

Use the following create-launch-template command to create a launch template that specifies that the Capacity Reservation targets a specific resource group. Replace the sample value for --launch-template-name. Replace c5.xlarge with the instance type that you used in the Capacity Reservation and ami-0123456789EXAMPLE with the ID of the AMI that you want to use. Replace arn:aws:resource-groups:region:account-id:group/my-cr-group with the ARN of the resource group that you created in the beginning of the previous section.

aws ec2 create-launch-template \
    --launch-template-name my-launch-template \
    --launch-template-data \
        '{"InstanceType": "c5.xlarge",
          "ImageId": "ami-0123456789EXAMPLE",
          "CapacityReservationSpecification": 
            {"CapacityReservationTarget": 
                { "CapacityReservationResourceGroupArn": "arn:aws:resource-groups:region:account-id:group/my-cr-group" }
            }
        }'

The following is an example response.

{
    "LaunchTemplate": {
        "LaunchTemplateId": "lt-0dd77bd41dEXAMPLE",
        "LaunchTemplateName": "my-launch-template",
        "CreateTime": "2023-07-26T21:42:48+00:00",
        "CreatedBy": "arn:aws:iam::123456789012:user/Bob",
        "DefaultVersionNumber": 1,
        "LatestVersionNumber": 1
    }
}

Step 4: Create an Auto Scaling group

Console

Create your Auto Scaling group as you usually do, but when you choose your VPC subnets, choose a subnet from each Availability Zone that matches the targeted Capacity Reservations you created. Then, when your Auto Scaling group launches an On-Demand Instance in one of these Availability Zones, the instance will be run in the reserved capacity for that Availability Zone. If the resource group runs out of Capacity Reservations before your desired capacity is fulfilled, we launch anything beyond the reserved capacity as regular On-Demand capacity.

To create a simple Auto Scaling group

  1. Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/, and choose Auto Scaling Groups from the navigation pane.

  2. On the navigation bar at the top of the screen, choose the same AWS Region that you used when you created the launch template.

  3. Choose Create an Auto Scaling group.

  4. On the Choose launch template or configuration page, for Auto Scaling group name, enter a name for your Auto Scaling group.

  5. For Launch template, choose an existing launch template.

  6. For Launch template version, choose whether the Auto Scaling group uses the default, the latest, or a specific version of the launch template when scaling out.

  7. On the Choose instance launch options page, skip the Instance type requirements section to use the EC2 instance type that is specified in the launch template.

  8. Under Network, for VPC, choose a VPC. The Auto Scaling group must be created in the same VPC as the security group you specified in your launch template. If you didn't specify a security group in your launch template, you can choose any VPC that has subnets in the same Availability Zones as your Capacity Reservations.

  9. For Availability Zones and subnets, choose subnets from each Availability Zone that you want to include, based on which Availability Zones your Capacity Reservations are in.

  10. Choose Next twice.

  11. On the Configure group size and scaling policies page, for Desired capacity, enter the initial number of instances to launch. When you change this number to a value outside of the minimum or maximum capacity limits, you must update the values of Minimum capacity or Maximum capacity. For more information, see Set scaling limits for your Auto Scaling group.

  12. Choose Skip to review.

  13. On the Review page, choose Create Auto Scaling group.

AWS CLI
To create a simple Auto Scaling group

Use the following create-auto-scaling-group command and specify the name and version of your launch template as the value for the --launch-template option. Replace the sample values for --auto-scaling-group-name, --min-size, --max-size, and --vpc-zone-identifier.

For the --availability-zones option, specify the Availability Zones that you created Capacity Reservations for. For example, if your Capacity Reservations specify the us-east-1a and us-east-1b Availability Zones, then you must create your Auto Scaling group in the same zones. Then, when your Auto Scaling group launches an On-Demand Instance in one of these Availability Zones, the instance will be run in the reserved capacity for that Availability Zone. If the resource group runs out of Capacity Reservations before your desired capacity is fulfilled, we launch anything beyond the reserved capacity as regular On-Demand capacity.

aws autoscaling create-auto-scaling-group \
    --auto-scaling-group-name my-asg \
    --launch-template LaunchTemplateName=my-launch-template,Version='1' \
    --min-size 6 \
    --max-size 6 \
    --vpc-zone-identifier "subnet-5f46ec3b,subnet-0ecac448" \
    --availability-zones us-east-1a us-east-1b

For an example implementation, see the AWS CloudFormation template in the following AWS samples GitHub repository: https://github.com/aws-samples/aws-auto-scaling-backed-by-on-demand-capacity-reservations/.

The following related topics can be helpful as you learn about Capacity Reservations.