autoScalingMetric

The metric your cluster will track in order to scale in and out. For example, CPU_UTILIZATION_PERCENTAGE is the average CPU usage across all the nodes in a cluster.