What problem does it solve?
Pinpoints where server-side latency and resource bottlenecks occur when HTTP endpoints are slow, CPU or memory spikes appear, or logs do not reveal a clear root cause, enabling targeted fixes before or after traffic increases.
Core Features & Use Cases
- Quick diagnostics: endpoint timing, PHP-FPM worker counts, process CPU/memory snapshots, and open MySQL connections to locate hot processes fast.
- Load testing & SLO validation: guidance and k6/ab examples to simulate traffic, measure P95 latency, and validate capacity before campaigns or releases.
- Deep profiling: temporary Xdebug profiling, analysis of Xdebug/trace outputs, and techniques for finding N+1 queries and slow MySQL statements using slow logs and pt-query-digest.
- Frontend/SSR and cache checks: Next.js SSR timing, systemd logs inspection, Redis hit-rate analysis, and live monitoring patterns during tests.
- Use Case: Before a paid acquisition push, run latency snapshots and a k6 smoke test to confirm P95 < 1s and identify whether PHP-FPM tuning, query optimization, or Redis caching is needed.
Quick Start
Use the performance-profiling skill to run endpoint latency checks, a k6 smoke test, and identify the top PHP or MySQL hotspots causing elevated P95 or CPU load.
