FlameGraph Visualization Tool

FlameGraph is a visualization tool for profiling and performance analysis, created by Brendan Gregg (Distinguished Engineer at Netflix and Intel) as part of his systems performance methodology work. The tool processes stack trace samples from various profiling sources and generates interactive SVG visualizations known as flame graphs, which have become the industry standard for visualizing CPU, memory, and off-CPU profiling data. Key features include: flame graph visualization (stack traces displayed as horizontal bars with width proportional to execution time, color-coded by function type, with y-axis showing call stack depth and x-axis showing sample population sorted alphabetically), multi-language profiling support (processing output from DTrace, perf, eBPF, SystemTap, bpftrace, Python, Java, Node.js, Go, PHP, Ruby, and Rust profilers via stackcollapse scripts), differential flame graphs (comparing two profiles side by side to identify performance regressions between code versions), CPU profiling (on-CPU flame graphs showing where CPU time is spent), off-CPU profiling (off-CPU flame graphs showing where threads are blocked, waiting on I/O, locks, or timers), memory profiling (allocation flame graphs showing heap allocation hotspots), and flame chart (time-ordered flame graphs preserving chronological execution order rather than alphabetical sorting), interactive SVG (zoomable and searchable SVG output with tooltips showing function details and sample counts), Perl-based generation (lightweight Perl scripts: flamegraph.pl, stackcollapse-perf.pl, stackcollapse-dtrace.pl, and language-specific stackcollapse variants), and deployment in production (widely used at Netflix, Meta, Google, AWS, and major tech companies for continuous profiling and performance investigation).