COST03-BP06 Allocate costs based on workload metrics - Cost Optimization Pillar

COST03-BP06 Allocate costs based on workload metrics

Allocate the workload's costs based on usage metrics or business outcomes to measure workload cost efficiency. Implement a process to analyze the cost and usage data with analytics services, which can provide insight and charge back capability.

Level of risk exposed if this best practice is not established: Low

Implementation guidance

Cost optimization means delivering business outcomes at the lowest price point, which can only be achieved by allocating workload costs based on workload metrics (measured by workload efficiency). Monitor the defined workload metrics through log files or other application monitoring. Combine this data with the workload’s costs, which can be obtained by looking at costs with a specific tag value or account ID. Perform this analysis at the hourly level. Your efficiency typically changes if you have static cost components (for example, a backend database running permanently) with a varying request rate (for example, usage peaks at nine in the morning to five in the evening, with few requests at night). Understanding the relationship between the static and variable costs helps you focus your optimization activities.

Creating workload metrics for shared resources may be challenging compared to resources like containerized applications on Amazon Elastic Container Service (Amazon ECS) and Amazon API Gateway. However, there are certain ways you can categorize usage and track cost. If you need to track Amazon ECS and AWS Batch shared resources, you can enable split cost allocation data in AWS Cost Explorer. With split cost allocation data, you can understand and optimize the cost and usage of your containerized applications and allocate application costs back to individual business entities based on how shared compute and memory resources are consumed.

Implementation steps

  • Allocate costs to workload metrics: Using the defined metrics and configured tags, create a metric that combines the workload output and workload cost. Use analytics services such as Amazon Athena and Amazon QuickSight to create an efficiency dashboard for the overall workload and any components.


Related documents:

Related examples: