Question

A SysOps administrator manages a fleet of 40 m5.2xlarge EC2 instances running a web application. Over the past 30 days, CloudWatch shows average CPU utilization of 8% and peak CPU of 15%. The team wants to reduce compute costs while maintaining current performance. Before recommending a smaller instance size, the administrator wants to confirm that the workload is not memory-bound. What is the correct first step to obtain the data needed?

Accepted Answer

Install and configure the CloudWatch agent on the instances to collect memory utilization metrics, since memory is not reported by default EC2 metrics.

Answer

Review the default CloudWatch MemoryUtilization metric in the EC2 namespace to determine average RAM usage.

Answer

Enable detailed monitoring on the instances to expose memory metrics at 1-minute granularity.

Answer

Open AWS Compute Optimizer and rely solely on its CPU-based recommendation without further memory analysis.