Create a step scaling policy for Application Auto Scaling using the AWS CLI - Application Auto Scaling
Step 1: Register a scalable targetStep 2: Create a step scaling policyStep 3: Create an alarm that invokes a scaling policy

Create a step scaling policy for Application Auto Scaling using the AWS CLI

This example uses AWS CLI commands to create a step scaling policy for an Amazon ECS service. For a different scalable target, specify its namespace in --service-namespace, its scalable dimension in --scalable-dimension, and its resource ID in --resource-id.

When using the AWS CLI, remember that your commands run in the AWS Region configured for your profile. If you want to run the commands in a different Region, either change the default Region for your profile, or use the --region parameter with the command.

Step 1: Register a scalable target

If you haven't already done so, register the scalable target. Use the register-scalable-target command to register a specific resource in the target service as a scalable target. The following example registers an Amazon ECS service with Application Auto Scaling. Application Auto Scaling can scale the number of tasks at a minimum of 2 tasks and a maximum of 10. Replace each user input placeholder with your own information.

Linux, macOS, or Unix

aws application-autoscaling register-scalable-target --service-namespace ecs \
  --scalable-dimension ecs:service:DesiredCount \
  --resource-id service/my-cluster/my-service \
  --min-capacity 2 --max-capacity 10

Windows

aws application-autoscaling register-scalable-target --service-namespace ecs ^
  --scalable-dimension ecs:service:DesiredCount ^
  --resource-id service/my-cluster/my-service ^
  --min-capacity 2 --max-capacity 10
Output

If successful, this command returns the ARN of the scalable target. The following is example output.

{
    "ScalableTargetARN": "arn:aws:application-autoscaling:region:account-id:scalable-target/1234abcd56ab78cd901ef1234567890ab123"
}

Step 2: Create a step scaling policy

To create a step scaling policy for your scalable target, you can use the following examples to help you get started.

Scale out
To create a step scaling policy for scale out (increase capacity)

  1. Use the following cat command to store a step scaling policy configuration in a JSON file named config.json in your home directory. The following is an example configuration with an adjustment type of PercentChangeInCapacity that increases the capacity of the scalable target based on the following step adjustments (assuming a CloudWatch alarm threshold of 70):

    • Increase capacity by 10 percent when the value of the metric is greater than or equal to 70 but less than 85

    • Increase capacity by 20 percent when the value of the metric is greater than or equal to 85 but less than 95

    • Increase capacity by 30 percent when the value of the metric is greater than or equal to 95

    $ cat ~/config.json
{
  "AdjustmentType": "PercentChangeInCapacity",
  "MetricAggregationType": "Average",
  "Cooldown": 60,
  "MinAdjustmentMagnitude": 1,
  "StepAdjustments": [ 
    {
      "MetricIntervalLowerBound": 0.0,
      "MetricIntervalUpperBound": 15.0,
      "ScalingAdjustment": 10
    },
    {
      "MetricIntervalLowerBound": 15.0,
      "MetricIntervalUpperBound": 25.0,
      "ScalingAdjustment": 20
    },
    {
      "MetricIntervalLowerBound": 25.0,
      "ScalingAdjustment": 30
    }
  ]
}

    For more information, see StepScalingPolicyConfiguration in the Application Auto Scaling API Reference.

  2. Use the following put-scaling-policy command, along with the config.json file that you created, to create a scaling policy named my-step-scaling-policy.

    Linux, macOS, or Unix

    aws application-autoscaling put-scaling-policy --service-namespace ecs \
  --scalable-dimension ecs:service:DesiredCount \
  --resource-id service/my-cluster/my-service \
  --policy-name my-step-scaling-policy --policy-type StepScaling \
  --step-scaling-policy-configuration file://config.json

    Windows

    aws application-autoscaling put-scaling-policy --service-namespace ecs ^
  --scalable-dimension ecs:service:DesiredCount ^
  --resource-id service/my-cluster/my-service ^
  --policy-name my-step-scaling-policy --policy-type StepScaling ^
  --step-scaling-policy-configuration file://config.json
    Output

    The output includes the ARN that serves as a unique name for the policy. You need it to create a CloudWatch alarm for your policy. The following is example output.

    {
    "PolicyARN": "arn:aws:autoscaling:region:123456789012:scalingPolicy:ac542982-cbeb-4294-891c-a5a941dfa787:resource/ecs/service/my-cluster/my-service:policyName/my-step-scaling-policy"
}
Scale in
To create a step scaling policy for scale in (decrease capacity)

  1. Use the following cat command to store a step scaling policy configuration in a JSON file named config.json in your home directory. The following is an example configuration with an adjustment type of ChangeInCapacity that decreases the capacity of the scalable target based on the following step adjustments (assuming a CloudWatch alarm threshold of 50):

    • Decrease capacity by 1 when the value of the metric is less than or equal to 50 but greater than 40

    • Decrease capacity by 2 when the value of the metric is less than or equal to 40 but greater than 30

    • Decrease capacity by 3 when the value of the metric is less than or equal to 30

    $ cat ~/config.json
{
  "AdjustmentType": "ChangeInCapacity",
  "MetricAggregationType": "Average",
  "Cooldown": 60,
  "StepAdjustments": [ 
    {
      "MetricIntervalUpperBound": 0.0,
      "MetricIntervalLowerBound": -10.0,
      "ScalingAdjustment": -1
    },
    {
      "MetricIntervalUpperBound": -10.0,
      "MetricIntervalLowerBound": -20.0,
      "ScalingAdjustment": -2
    },
    {
      "MetricIntervalUpperBound": -20.0,
      "ScalingAdjustment": -3
    }
  ]
}

    For more information, see StepScalingPolicyConfiguration in the Application Auto Scaling API Reference.

  2. Use the following put-scaling-policy command, along with the config.json file that you created, to create a scaling policy named my-step-scaling-policy.

    Linux, macOS, or Unix

    aws application-autoscaling put-scaling-policy --service-namespace ecs \
  --scalable-dimension ecs:service:DesiredCount \
  --resource-id service/my-cluster/my-service \
  --policy-name my-step-scaling-policy --policy-type StepScaling \
  --step-scaling-policy-configuration file://config.json

    Windows

    aws application-autoscaling put-scaling-policy --service-namespace ecs ^
  --scalable-dimension ecs:service:DesiredCount ^
  --resource-id service/my-cluster/my-service ^
  --policy-name my-step-scaling-policy --policy-type StepScaling ^
  --step-scaling-policy-configuration file://config.json
    Output

    The output includes the ARN that serves as a unique name for the policy. You need this ARN to create a CloudWatch alarm for your policy. The following is example output.

    {
    "PolicyARN": "arn:aws:autoscaling:region:123456789012:scalingPolicy:ac542982-cbeb-4294-891c-a5a941dfa787:resource/ecs/service/my-cluster/my-service:policyName/my-step-scaling-policy"
}

Step 3: Create an alarm that invokes a scaling policy

Finally, use the following CloudWatch put-metric-alarm command to create an alarm to use with your step scaling policy. In this example, you have an alarm based on average CPU utilization. The alarm is configured to be in an ALARM state if it reaches a threshold of 70 percent for at least two consecutive evaluation periods of 60 seconds. To specify a different CloudWatch metric or use your own custom metric, specify its name in --metric-name and its namespace in --namespace.

Linux, macOS, or Unix

aws cloudwatch put-metric-alarm --alarm-name Step-Scaling-AlarmHigh-ECS:service/my-cluster/my-service \
  --metric-name CPUUtilization --namespace AWS/ECS --statistic Average \
  --period 60 --evaluation-periods 2 --threshold 70 \
  --comparison-operator GreaterThanOrEqualToThreshold \
  --dimensions Name=ClusterName,Value=default Name=ServiceName,Value=sample-app-service \
  --alarm-actions PolicyARN

Windows

aws cloudwatch put-metric-alarm --alarm-name Step-Scaling-AlarmHigh-ECS:service/my-cluster/my-service ^
  --metric-name CPUUtilization --namespace AWS/ECS --statistic Average ^
  --period 60 --evaluation-periods 2 --threshold 70 ^
  --comparison-operator GreaterThanOrEqualToThreshold ^
  --dimensions Name=ClusterName,Value=default Name=ServiceName,Value=sample-app-service ^
  --alarm-actions PolicyARN