Description

This policy identifies AWS ElastiCache for Memcached Clusters that appear to be underutilized based on their average CPU utilization over the past 30 days.

Rationale

Operating underutilized ElastiCache clusters results in unnecessary cloud costs. By identifying and rightsizing these clusters, you can reduce expenses while maintaining application performance.

Impact

Before downsizing or decommissioning a cluster, verify that it is not supporting workloads with infrequent but high CPU demands. Premature rightsizing may introduce performance bottlenecks during peak usage periods.

Audit

This policy flags an AWS ElastiCache Cluster for Memcached as INCOMPLIANT if its CPU Utilization, Average, % metric is below 40% over the last 30 days.

ElastiCache Clusters for Redis are marked as INAPPLICABLE.

Clusters created within the last 30 days or not in an available Status are also marked as INAPPLICABLE.