Examkin

Question

Difficulty: MediumAWS Compute Optimizer and Right-Sizing

A SysOps Administrator is reviewing optimization recommendations in AWS Compute Optimizer for a fleet of EC2 instances hosting a memory-intensive batch processing application. The dashboard lists these instances as over-provisioned based on CPU utilization, but memory metrics are missing from the analysis, leading to potentially inaccurate right-sizing recommendations. Which action must the administrator take to enable AWS Compute Optimizer to ingest memory utilization metrics for these EC2 instances?

  1. Install and configure the unified CloudWatch agent on the EC2 instances to publish OS-level memory metrics to Amazon CloudWatch under the CWAgent namespace.Answer
  2. B
    Enable CloudWatch detailed monitoring on the EC2 instances to increase the reporting frequency of CPU and memory metrics to one-minute intervals.
  3. C
    Configure an AWS Budgets alert to monitor instance memory utilization and send a notification when usage exceeds 80%.
  4. D
    Enable cost allocation tags in the Billing Console and use AWS Cost Explorer to analyze memory utilization patterns across instance families.

Answer

Install and configure the unified CloudWatch agent on the EC2 instances to publish OS-level memory metrics to Amazon CloudWatch under the CWAgent namespace.
The correct answer is to install and configure the unified CloudWatch agent on the EC2 instances to publish memory metrics to the default CWAgent namespace. Because memory utilization is an operating system-level metric, the underlying AWS hypervisor does not have visibility into it. The unified CloudWatch agent must be installed to collect memory metrics and send them to CloudWatch, where AWS Compute Optimizer can retrieve and ingest them to deliver memory-aware right-sizing recommendations.

Step-by-Step Solution

1
Deploy the unified CloudWatch agent on the target EC2 instances.
The agent is installed on the instances' operating systems.
Memory utilization is an OS-level metric that the hypervisor cannot access directly; an agent is required to collect this data.
2
Configure the CloudWatch agent configuration file to collect memory metrics and send them to the default namespace.
Memory metrics such as mem_used_percent are published to the CWAgent namespace in Amazon CloudWatch.
AWS Compute Optimizer automatically looks for memory metrics published under the default CWAgent namespace to generate recommendations.
3
Allow AWS Compute Optimizer up to 30 hours to ingest the new metrics and update the right-sizing recommendations.
Compute Optimizer includes memory utilization in its optimization analysis.
Compute Optimizer requires a minimum duration of metric history to update its recommendations.

Key Concept

AWS Compute Optimizer relies on the unified CloudWatch agent to analyze memory metrics because memory is an operating system-level metric not visible to the hypervisor by default.
Estimated Time:1m 30s
Rate this question

Topics

AWS Compute Optimizer and Right-Sizing

More questions from this topic