Performance Efficiency
- The question it answers
- Are we using computing resources efficiently, and do we keep that efficiency as demand and technology change?
- Design principles
- Democratize advanced technologies — consume them as managed services instead of building expertise.
- Go global in minutes — deploy to multiple Regions for lower latency.
- Use serverless architectures — no servers to manage or right-size.
- Experiment more often — cheap, fast tests of different instance types and configurations.
- Consider mechanical sympathy — pick the technology that actually matches the access pattern.
- Typical services
- Right-sized Amazon EC2 families, AWS Lambda, Amazon CloudFront, AWS Global Accelerator, ElastiCache, AWS Compute Optimizer, AWS Auto Scaling.
- Distinguish from
- Cost Optimization — the same right-sizing action serves both, but Performance Efficiency asks “is it fast enough?” and Cost Optimization asks “are we paying more than we need?”