A company uses AWS Organizations to manage its AWS accounts. The company needs a list of all its Amazon EC2 instances that have underutilized CPU or memory usage. The company also needs recommendation

Installing a third-party monitoring tool adds complexity and effort compared to AWS native solutions, and AWS Trusted Advisor provides general recommendations but is less specialized for EC2 sizing than AWS Compute Optimizer.

AWS Systems Manager can efficiently deploy the Amazon CloudWatch agent to all EC2 instances across an organization to collect detailed memory metrics, which are not captured by default. AWS Compute Optimizer then analyzes these metrics along with CPU utilization to provide insights into underutilized instances and specific downsizing recommendations across the entire organization with minimal manual effort.

While collecting metrics via Systems Manager and the CloudWatch agent is correct, creating custom EventBridge rules and Lambda functions for downsizing logic and recommendations is significantly more operational overhead than utilizing AWS Compute Optimizer, which provides this functionality natively.

CloudWatch alarms can identify instances below certain thresholds, but they do not provide specific downsizing recommendations for instance types, requiring manual analysis which contradicts the 'least effort' requirement.