What is a recommended logging strategy balancing observability and performance?

• A. Verbose logs everywhere
• B. Structured, leveled logs with sampling, avoid sensitive data, centralized storage
• C. Disable logging in production
• D. Use random log levels

Answer: (B)

Balancing observability and performance comes from capturing meaningful, structured data without flooding the system or exposing sensitive information. Structured logs provide a consistent schema so tools can easily query fields like timestamps, service names, request IDs, and status codes, enabling cross-component correlation. Leveled logging lets you control verbosity by environment and situation, keeping production quiet while allowing deeper detail when debugging. Sampling reduces log volume in high-traffic paths, giving representative insight without logging every event. Avoiding sensitive data in logs protects privacy and minimizes risk. Centralized storage with proper access controls makes it possible to search, analyze, and retain logs across services efficiently. Together, these practices deliver useful visibility while keeping performance overhead manageable.