Quickly remove some big performance bottlenecks with QuickPerf
summary
QuickPerf shows concrete fixes: it detects N+1 SQL patterns by failing tests that execute one expected select but run hundreds instead (example: a test that should run 1 becomes ~600 statements), it measures heap allocations and GC activity via JFR/Java Mission Control integration, and it provides JUnit extensions and annotations to declare limits and auto-fail when exceeded. Demonstrations include replacing per-item select loops with one join query and cutting allocation hotspots, with failing green tests turning red when limits are breached.
AI-written summary · about
chapters (10)
- 0:00QuickPerf introduction
- 1:32Performance problems origins
- 4:22QuickPerf demo with tests
- 7:49Inefficient SQL queries detection
- 10:50Optimizing queries with annotations
- 13:55Memory allocations analysis
- 17:08SQL parameter query checks
- 19:43Production memory measurements
- 22:14Utilisation de QuickPerf avec Java Mission Control
- 24:32QuickPerf feedback
related talks
35mEliminate Performance Bottlenecks Quickly with QuickPerf
50mPerformance Testing Java Applications
29mJava Performance Testing for Everyone
45mPerformance lab. The power of the perf tool!
39mCreating & Scaling Meaningful Performance Tests in CI
55mNext-Gen Testing Tools for Java
54mPerformance tests and other lies
02h 44mMeasure, don’t guess: benchmarking stories from the trenches