performance-analyst

To debug performance regressions, you analyze profiling artifacts, flamegraphs, and runtime metrics. This process identifies key performance hotspots, maps them to specific code paths, and recommends low-risk optimizations with expected impact to resolve the slowdown.

A sudden increase in web application response time is typically caused by code path bottlenecks. You investigate it by analyzing server logs, flamegraphs, and profiling data to pinpoint the exact functions causing the regression and suggest targeted optimizations.

Identifying performance hotspots from runtime metrics alone is limited; this analysis requires access to profiling logs and flamegraphs. Combining runtime metrics with these detailed profiling artifacts ensures accurate mapping of bottlenecks to specific code paths for effective optimization.

The best way to find low-risk code optimizations is by analyzing profiling artifacts and flamegraphs to detect hotspots. This approach maps performance regressions to exact code paths and calculates expected impact, ensuring recommended optimizations carry minimal risk.

Yes, you need to provide access to profiling logs and flamegraphs to analyze a performance regression. Supplying these profiling artifacts and runtime metrics is required for the skill to generate detailed analysis reports and recommend automated code improvements.