Study Guide: Optimizing Compute with AWS Compute Optimizer

This guide covers the core functionalities, metrics requirements, and strategic implementation of AWS Compute Optimizer, focusing on the SOA-C03 exam objectives for cost and performance optimization.

Learning Objectives

After studying this guide, you should be able to:

• Identify which AWS resource types are supported by Compute Optimizer.
• Differentiate between the free (default) and paid versions of the service.
• Explain the prerequisites for analyzing EC2 memory utilization.
• Evaluate recommendations to mitigate performance risks and reduce costs.
• Configure Compute Optimizer at the AWS Organization level.

Key Terms & Glossary

• Rightsizing: The process of matching instance types and sizes to your workload performance and capacity requirements at the lowest possible cost.
• Over-provisioned: A resource state where the capacity exceeds the demand, leading to unnecessary costs.
• Under-provisioned: A resource state where demand exceeds capacity, potentially causing performance bottlenecks or application failure.
• CloudWatch Agent: A software package installed on EC2 instances to collect system-level metrics (like memory utilization) that are not available by default to the hypervisor.
• Enhanced Infrastructure Metrics: A paid feature that extends the look-back period for data analysis from 14 days to 3 months.

The "Big Idea"

AWS Compute Optimizer shifts infrastructure management from reactive guessing to proactive, data-driven decisions. By leveraging machine learning (ML) to analyze historical CloudWatch metrics, it provides specific "rightsizing" recommendations. This ensures that you aren't paying for idle CPU cycles (Cost Optimization) while also ensuring your applications have enough