HNSUS02-BP01 Prioritize critical components
Break down your workload into individual components (for example, microservices, databases, and APIs). Use metrics like CPU utilization, request rates, and business impact to classify them as critical or non-critical.
Desired outcome: Resource allocation aligned with business value, minimizing waste in low-impact areas.
Level of risk exposed if this best practice is not established: Medium
Benefits of establishing this best practice:
-
Focuses optimization efforts on high-impact components
-
Reduces energy consumption
-
Maintains performance for mission-critical workloads
Implementation guidance
-
Map component dependencies and usage. For example, you can achieve this using AWS X-Ray.
-
Apply tags to categorize components (for example, business-critical, dev-test).
-
Right-size your resources during off-peak hours. For example, you can achieve this using AWS Auto Scaling.