Inside JMH: How the Microbenchmark Harness Defeats Dead Code
Understand jmh benchmark dead code elimination, the main trade-offs, and the practical checks to use before relying on it in practice.
Inside JFR Event Streaming: How the JVM Buffers Events Without
JFR streams events without halting the application because the emitter and the consumer don't share a mutable in-memory queue.
JFR vs async-profiler for production CPU profiling
For production CPU profiling on a JDK 21 or 25 box in 2026, default to async-profiler with --jfrsync : roughly 1% overhead at -i 10ms , no safepoint bias.