Flame Graph for Performance Profiling in Python, R, Julia and JavaScript

Source code in 15 libraries — Python: Altair, Bokeh, lets-plot, Matplotlib, Plotly, plotnine, Pygal, Seaborn; R: ggplot2; Julia: Makie.jl; JavaScript: Chart.js, D3.js, Apache ECharts, Highcharts, MUI X Charts.

A flame graph visualizes hierarchical call stack data from performance profiling, where each horizontal bar represents a function in the call stack and its width is proportional to the time (or samples) spent in that function. Stacks are layered bottom-to-top showing caller-to-callee relationships. Invented by Brendan Gregg, flame graphs are the standard visualization for identifying CPU bottlenecks and hot code paths across all major programming languages and profiling tools.

Flame Graph for Performance Profiling

Renders

Preview render from the Highcharts implementation.

Implementations