AWS Compute Optimizer Cheat Sheet

• A service that recommends optimal AWS resources to reduce costs and improve performance of your workloads.
• Uses machine learning to analyze historical utilization metrics.

• You can view findings and recommendations across AWS Regions and accounts.
• Generates recommendations for the following resources:
  • Amazon EC2 instances
  • Amazon EC2 Auto Scaling groups
  • Amazon EBS volumes
  • AWS Lambda functions

Features

• New EC2 instance type recommendations: Compute Optimizer now provides recommendations for the latest EC2 instance types, including C8gn, I8ge, M8i, R8i, and more.
• Expanded EC2 Auto Scaling group recommendations: Supports groups using G or P instance types, mixed instance types, scaling policies, and groups scaled down to a single instance.
• Aurora I/O-Optimized recommendations: Compute Optimizer now provides recommendations for Aurora DB cluster storage.
• RDS rightsizing recommendations: Supports Aurora MySQL, Aurora PostgreSQL, Amazon RDS MySQL, and RDS PostgreSQL.
• Rightsizing preferences for memory utilization headroom: Customize memory headroom settings for EC2 instance recommendations.
• Savings estimation mode: Allows Compute Optimizer to consider Savings Plans and Reserved Instances when estimating cost savings.
• GPU-based EC2 instance recommendations: Includes G4dn and P3 instance types.
• License recommendations for Microsoft SQL Server: Provides insights for commercial software licenses running on EC2.
• Tag filtering and inferred workload type filtering: Filter recommendations by tag key/value and inferred workload types.
• EBS volume support: Now includes HDD st1, sc1, and Provisioned IOPS SSD io2 Block Express.
• EC2 suspension workloads: Combines intermittent utilization data to generate EC2 recommendations.
• ECS services on Fargate recommendations: Provides rightsizing recommendations for ECS services on Fargate.
• External metrics ingestion: Ingest external EC2 memory utilization metrics from supported observability tools for more accurate recommendations.

Concepts

• Recommendation preferences are features that you can activate to enhance or augment the recommendations generated for your resources.
  • Enhanced infrastructure metrics – extends the look-back period for utilization metrics analysis to three months.
  • Inferred workload type – identify the effort to migrate workloads from x86-based to Arm-based AWS Graviton instances types.
  • AWS Graviton-based instance recommendations – provides price and performance impact.
• Delegated Administrator
  • You can delegate a member account in your organization as an administrator.
  • There can only be one delegated administrator per organization.
  • The delegated administrator can do the following:
    • Get and export recommendations
    • Get projected utilization metrics
    • Set recommendation preferences
    • Set member account opt-in status
  • You can use AWS console and CLI to register and remove an account as a delegated administrator.
• The dashboard allows you to view the optimization findings for your AWS resources across all AWS Regions.
• You can also export and share your recommendations to an Amazon S3 bucket.

AWS Compute Optimizer Pricing

• You are charged for the enhanced infrastructure metrics per resource.
• You are charged based on the number of hours the resource runs.

AWS Compute Optimizer Cheat Sheet References:

https://aws.amazon.com/compute-optimizer/
https://docs.aws.amazon.com/compute-optimizer/latest/ug/what-is-compute-optimizer.html