Keeping Your Java Hot.
summary
Compares concrete fixes for JVM warmup: Graal native image gives immediate full-speed start but loses speculative optimizations; ReadyNow / Project Liden and platform Prime capture JIT profiles and cached compiled code for reuse; a cloud-native JIT-as-a-Service decouples compilation, offloads CPU and lets one-time heavy optimizations be reused; CRaC checkpoint/restore snapshots heap+JIT state and cuts First-transaction from ~3.9s to 38ms (Spring), ~1s to 46ms (Micron), ~980ms to 33ms (Quarkus).
AI-written summary · about
chapters (6)
- 0:00JVM warmup challenges
- 10:26AOT compilation explained
- 27:52Storing JIT data
- 35:03Decoupling JIT compiler
- 47:06Saving app state
- 55:41JVM warmup solutions
related talks
46mKeeping Your Java Hot by Solving the JVM Warmup Problem
54mKeeping Your Java Hot by Solving the JVM Warmup Problem
56mKeeping Your #Java Hot by Solving the #JVM Warmup Problem
27mHow to Eliminate Java Performance Warmup
06mSimple Tweaks to Get the Most Out of your JVM
42mTechniques for a faster JVM start-up
35mJava at Speed: Building a Better JVM
50mAOT or JIT: Faster Startup or Faster Code?