Reading Flame Graphs: From a Wide Plateau to a Line of Code
How to interpret width versus height, distinguish on-CPU from off-CPU, and use allocation flames to catch the copy you cannot see in the source.
Tag
5 articles · 16 min of reading
How to interpret width versus height, distinguish on-CPU from off-CPU, and use allocation flames to catch the copy you cannot see in the source.
Always-on JFR: which event templates belong in prod, how to dump recordings on a stall, and what JFR will not tell you that async-profiler will.
cycles, instructions, IPC, cache misses, and stalled-cycles: using Linux perf before you rewrite a method the profiler barely saw.
Guessing at performance problems wastes time twice: once guessing, once confirming the guess was wrong. Here's how to actually see what the JVM is doing.
The fastest way to waste a sprint is optimizing code that was never the bottleneck — a repeatable workflow for measuring before you touch anything.