Performance optimisation: how do I go about it?
summary
A small CLI URL checker named cigar profiles from ~3.0s down to ~0.3s after diagnosing an expensive array_map closure that computes PI and removing it; workflow shows XHProf, Xdebug, QCacheGrind and meminfo for profiling, call graphs and heap inspection, with concrete numbers and before/after traces. The process emphasizes measure then change: record a baseline, open profiler output, find the big empty squares, eliminate unused objects and confirm memory and timing improvements.
AI-written summary · about
chapters (7)
- 0:00Start
- 13:22MICRO OPTIMISATIONS
- 18:18OFF THE SHELF
- 20:32TOOLS FOR PHP
- 22:23INSTALL A PROFILER
- 26:29BASELINE FIRST
- 46:53SILVER SPONSOR
related talks
52mPerformance optimisation: how do I go about it?
50mPerformance optimization: how do I go about it?
29mHow to Performance
01h 03mThe performance loop—A practical guide to profiling and benchmarking
28mPerformance Profiling for V8
01h 03mApplication Performance Optimisation in Practice
02h 33mMeasure, don't Guess - How to find out if and where to optimize
39mProfiling and Benchmarking 101