Amazon CloudWatch metrics for Amazon EC2 Auto Scaling - Amazon EC2 Auto Scaling
Auto Scaling group metricsDimensions for Auto Scaling group metricsPredictive scaling metrics and dimensionsEnable Auto Scaling group metrics (console)Enable Auto Scaling group metrics (AWS CLI)

Amazon CloudWatch metrics for Amazon EC2 Auto Scaling

Amazon EC2 Auto Scaling publishes the following metrics in the AWS/AutoScaling namespace. The actual Auto Scaling group metrics made available will depend on whether you have group metrics enabled, and which group metrics you enabled. Group metrics are available at one-minute granularity at no additional charge, but you must enable them.

When you enable Auto Scaling group metrics, Amazon EC2 Auto Scaling sends sampled data to CloudWatch every minute on a best-effort basis. In rare cases when CloudWatch experiences a service disruption, data isn't backfilled to fill gaps in group metric history.

Auto Scaling group metrics

With these metrics, you get nearly continuous visibility into the history of your Auto Scaling group, such as changes in the size of the group over time.

Metric Description
GroupMinSize

The minimum size of the Auto Scaling group.

Reporting criteria: Reported if metrics collection is enabled.
GroupMaxSize

The maximum size of the Auto Scaling group.

Reporting criteria: Reported if metrics collection is enabled.
GroupDesiredCapacity

The number of instances that the Auto Scaling group attempts to maintain.

Reporting criteria: Reported if metrics collection is enabled.
GroupInServiceInstances

The number of instances that are running as part of the Auto Scaling group. This metric does not include instances that are pending or terminating.

Reporting criteria: Reported if metrics collection is enabled.
GroupPendingInstances

The number of instances that are pending. A pending instance is not yet in service. This metric does not include instances that are in service or terminating.

Reporting criteria: Reported if metrics collection is enabled.
GroupStandbyInstances

The number of instances that are in a Standby state. Instances in this state are still running but are not actively in service.

Reporting criteria: Reported if metrics collection is enabled.
GroupTerminatingInstances

The number of instances that are in the process of terminating. This metric does not include instances that are in service or pending.

Reporting criteria: Reported if metrics collection is enabled.
GroupTotalInstances

The total number of instances in the Auto Scaling group. This metric identifies the number of instances that are in service, pending, and terminating.

Reporting criteria: Reported if metrics collection is enabled.

When you configure a mixed instances group to measure its desired capacity in different units, such as by assigning weights based on the vCPU count of each instance type, the following metrics count the number of units used by your Auto Scaling group. If you did not configure a mixed instances group to measure its desired capacity in different units, then the following metrics are populated, but are equal to the metrics that are defined in the previous table. For more information, see Setup overview for creating a mixed instances group.

Metric Description
GroupInServiceCapacity

The number of capacity units that are running as part of the Auto Scaling group.

Reporting criteria: Reported if metrics collection is enabled.
GroupPendingCapacity

The number of capacity units that are pending.

Reporting criteria: Reported if metrics collection is enabled.
GroupStandbyCapacity

The number of capacity units that are in a Standby state.

Reporting criteria: Reported if metrics collection is enabled.
GroupTerminatingCapacity

The number of capacity units that are in the process of terminating.

Reporting criteria: Reported if metrics collection is enabled.
GroupTotalCapacity

The total number of capacity units in the Auto Scaling group.

Reporting criteria: Reported if metrics collection is enabled.

Amazon EC2 Auto Scaling also reports the following metrics for Auto Scaling groups that have a warm pool. For more information, see Decrease latency for applications with long boot times using warm pools.

Metric Description
WarmPoolMinSize

The minimum size of the warm pool.

Reporting criteria: Reported if metrics collection is enabled.
WarmPoolDesiredCapacity

The amount of capacity that Amazon EC2 Auto Scaling attempts to maintain in the warm pool.

This is equivalent to the maximum size of the Auto Scaling group minus its desired capacity, or, if set, as the maximum prepared capacity of the Auto Scaling group minus its desired capacity.

However, when the minimum size of the warm pool is equal to or greater than the difference between the maximum size (or, if set, the maximum prepared capacity) and the desired capacity of the Auto Scaling group, then the warm pool desired capacity will be equivalent to the WarmPoolMinSize.

Reporting criteria: Reported if metrics collection is enabled.
WarmPoolPendingCapacity

The amount of capacity in the warm pool that is pending. This metric does not include instances that are running, stopped, or terminating.

Reporting criteria: Reported if metrics collection is enabled.
WarmPoolTerminatingCapacity

The amount of capacity in the warm pool that is in the process of terminating. This metric does not include instances that are running, stopped, or pending.

Reporting criteria: Reported if metrics collection is enabled.
WarmPoolWarmedCapacity

The amount of capacity available to enter the Auto Scaling group during scale out. This metric does not include instances that are pending or terminating.

Reporting criteria: Reported if metrics collection is enabled.
WarmPoolTotalCapacity

The total capacity of the warm pool, including instances that are running, stopped, pending, or terminating.

Reporting criteria: Reported if metrics collection is enabled.
GroupAndWarmPoolDesiredCapacity

The desired capacity of the Auto Scaling group and the warm pool combined.

Reporting criteria: Reported if metrics collection is enabled.
GroupAndWarmPoolTotalCapacity

The total capacity of the Auto Scaling group and the warm pool combined. This includes instances that are running, stopped, pending, terminating, or in service.

Reporting criteria: Reported if metrics collection is enabled.

Dimensions for Auto Scaling group metrics

You can use the following dimensions to refine the metrics listed in the previous tables.

Dimension Description
AutoScalingGroupName

Filters on the name of an Auto Scaling group.

Predictive scaling metrics and dimensions

The AWS/AutoScaling namespace includes the following metrics for predictive scaling.

Metrics are available with a resolution of one hour.

You can evaluate forecast accuracy by comparing forecasted values with actual values. For more information about evaluating forecast accuracy using these metrics, see Monitor predictive scaling metrics with CloudWatch.

Metric Description Dimensions
PredictiveScalingLoadForecast

The amount of load that's anticipated to be generated by your application.

The Average, Minimum, and Maximum statistics are useful, but the Sum statistic is not.

Reporting criteria: Reported after the initial forecast is created.

 AutoScalingGroupName, PolicyName, PairIndex
PredictiveScalingCapacityForecast

The anticipated amount of capacity needed to meet application demand. This is based on the load forecast and target utilization level at which you want to maintain your Auto Scaling instances.

The Average, Minimum, and Maximum statistics are useful, but the Sum statistic is not.

Reporting criteria: Reported after the initial forecast is created.

 AutoScalingGroupName, PolicyName
PredictiveScalingMetricPairCorrelation

The correlation between the scaling metric and the per-instance average of the load metric. Predictive scaling assumes high correlation. Therefore, if you observe low value for this metric, it's better not to use a metric pair.

The Average, Minimum, and Maximum statistics are useful, but the Sum statistic is not.

Reporting criteria: Reported after the initial forecast is created.

 AutoScalingGroupName, PolicyName, PairIndex
Note

The PairIndex dimension returns information associated with the index of the load-scaling metric pair as assigned by Amazon EC2 Auto Scaling. Currently, the only valid value is 0.

Enable Auto Scaling group metrics (console)

To enable group metrics

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

  2. Select the check box next to your Auto Scaling group.

    A split pane opens up in the bottom of the page.

  3. On the Monitoring tab, select the Auto Scaling group metrics collection, Enable check box located at the top of the page under Auto Scaling.

To disable group metrics

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

  2. Select your Auto Scaling group.

  3. On the Monitoring tab, clear the Auto Scaling group metrics collection, Enable check box.

Enable Auto Scaling group metrics (AWS CLI)

To enable Auto Scaling group metrics

Enable one or more group metrics by using the enable-metrics-collection command. For example, the following command enables a single metric for the specified Auto Scaling group.

aws autoscaling enable-metrics-collection --auto-scaling-group-name my-asg \
  --metrics GroupDesiredCapacity --granularity "1Minute"

If you omit the --metrics option, all metrics are enabled.

aws autoscaling enable-metrics-collection --auto-scaling-group-name my-asg \
  --granularity "1Minute"
To disable Auto Scaling group metrics

Use the disable-metrics-collection command to disable all group metrics.

aws autoscaling disable-metrics-collection --auto-scaling-group-name my-asg