Use AWS Compute Optimizer to examine individual resources and decide whether to resize, replace, stop, or remove them using utilization and performance context. Use AWS Cost Optimization Hub to bring cost-saving opportunities together across AWS accounts and Regions, then filter and prioritize them—including rightsizing and commitment-based savings. Use both when you need detailed resource recommendations and a portfolio-wide view.
How the two services differ
The simplest distinction is resource-level analysis versus portfolio-level consolidation. Compute Optimizer analyzes resource configuration and utilization, presents recommendations with recent and projected utilization context, and helps assess price-performance trade-offs. Cost Optimization Hub aggregates cost optimization recommendations, accounts for AWS commercial terms, and helps teams compare and prioritize opportunities across accounts and Regions.
They are not mutually exclusive tools. AWS connects Compute Optimizer rightsizing and idle-resource recommendations to the Hub’s aggregation layer. AWS also says that after Cost Optimization Hub is enabled, its estimated monthly savings appear in Compute Optimizer.
Choose based on the job you need to do
| Your main need | Start with | Why |
|---|---|---|
| Inspect utilization and choose a right-sized resource configuration | Compute Optimizer | It shows resource findings, recommendation options, and utilization and projected-utilization context. |
| Identify idle resources and resource-level actions | Compute Optimizer | It analyzes resource specifications and utilization for supported resource types. |
| Aggregate opportunities across AWS accounts and Regions | Cost Optimization Hub | Consolidation and prioritization across those scopes are the Hub’s core purpose. |
| Compare rightsizing with Savings Plans or Reserved Instances | Cost Optimization Hub | Its opportunity categories include those commitments, and its savings estimates account for AWS commercial terms. |
| Review resource details and performance context, then rank a wider opportunity list | Both | Use Compute Optimizer for individual-resource analysis and the Hub for cross-account aggregation and prioritization. |
What each service covers
AWS Compute Optimizer
Compute Optimizer provides recommendations for supported resources including EC2, Auto Scaling, EBS, Lambda, ECS on Fargate, commercial software licenses, Aurora and RDS, NAT Gateway, DynamoDB, ElastiCache, MemoryDB, DocumentDB, WorkSpaces, and SageMaker. Its scope is not simply “find cheaper instances”: recommendations may involve changing configuration or addressing an idle resource. See AWS’s Compute Optimizer overview for the current supported-resource list and service details.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →#1 Best Overall
AWS Cost Optimization Hub
The Hub consolidates recommendations such as rightsizing, deleting idle resources, Savings Plans, and Reserved Instances. AWS says it incorporates commercial terms and aggregates and deduplicates savings across related opportunities, which helps avoid treating overlapping recommendations as independent savings. It supports filtering, sorting, and grouping to help prioritize the consolidated list. See AWS’s Cost Optimization Hub documentation for current coverage.
The services’ supported-resource lists and recommendation categories can change. Check the live AWS documentation before relying on coverage for a particular resource or commitment type.
Requirements and timing to account for
- Opt in: Compute Optimizer does not analyze resources until it is enabled. AWS states, “You must opt in to have Compute Optimizer analyze your AWS resources.”
- EC2 and Auto Scaling data: AWS requires at least 30 hours of CloudWatch metric data in the previous 14 days for recommendations for EC2 instances and Auto Scaling groups.
- Time to recommendations: After analysis has sufficient data, recommendations may take up to 24 hours to appear.
- Lambda eligibility: The documented Lambda memory-size recommendation condition includes configured memory of 1,792 MB or less and at least 50 invocations during the previous 14 days. These are eligibility thresholds, not a guarantee that every function meeting them will receive a recommendation.
- Lookback period: Compute Optimizer’s default analysis uses 14 days. Enhanced infrastructure metrics can extend the lookback to 93 days for selected resources; AWS identifies enhanced infrastructure metrics as a paid feature.
- Organization scope: Both services document cross-Region findings. A consolidated view across an organization requires opt-in by the organization’s management account.
Resource-specific conditions apply, so an enabled service may still have no recommendation for a given resource. AWS maintains the details in its Compute Optimizer resource requirements.
How to interpret savings and performance information
Compute Optimizer’s EC2 recommendations refresh daily and may present up to three options with utilization graphs and performance-risk information. A savings figure is an estimate, not a guaranteed reduction in the bill. Its pricing assumptions vary: when Cost Optimization Hub is enabled in Cost Explorer, Compute Optimizer uses Hub data that includes account-specific discounts; otherwise, AWS says it uses Cost Explorer and On-Demand pricing information. Review the assumptions and settings behind a recommendation in AWS’s EC2 instance recommendation guidance.
Rank #3
The Hub is useful when comparing the combined opportunity set, particularly where commercial terms and commitments affect the cost picture. Neither a ranked recommendation nor an estimated monthly saving substitutes for checking workload requirements, purchase commitments, and the pricing basis that applies to your account.
A practical workflow
- Enable the analysis you need. Opt in to Compute Optimizer; if you need organization-wide visibility, have the management account opt in. Enable Cost Optimization Hub for consolidated opportunity analysis.
- Allow for data and analysis. Check that each resource meets its metric and service-specific eligibility conditions, and allow recommendations time to appear.
- Investigate resource-level recommendations. In Compute Optimizer, review utilization history, projected utilization, recommendation options, performance risk, and the savings assumptions before changing a resource.
- Prioritize across the estate. In Cost Optimization Hub, filter, sort, and group the cross-account opportunities; account for overlaps and commitment recommendations when deciding what to act on first.
- Validate before implementing. Confirm that the selected change fits workload needs and account-specific pricing or commitments, then monitor the outcome.
Which should you use?
For an immediate question about how a specific AWS resource is performing and whether its configuration should change, start with Compute Optimizer. For a team deciding which savings opportunities to tackle across accounts and Regions—or comparing rightsizing with Savings Plans and Reserved Instances—start with Cost Optimization Hub. When both levels matter, use the Hub to prioritize and Compute Optimizer to inspect the resource-level evidence.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
