Running Java on ARM - is it worth it?
summary
Shows measured instance-level comparisons across two clouds and cites an independent benchmark used as a baseline, concluding typical savings of 15–20% cost savings on ARM instances; uses Graviton/Ampere/Axon platforms, Docker/QEMU for cross-building, and JVM feature notes: JDK 11+ required for performant ports, Vector API and HotSpot work well, GraalVM native-image supports Linux and macOS but no Windows native image support, and heavy JNI users must recompile JNI deps or expect trouble.
AI-written summary · about
chapters (11)
- 0:00ARM CPUs overview
- 2:37What ARM CPUs are
- 9:53ARM architecture history
- 15:40ARM in cloud dominance
- 18:09Java on ARM history
- 23:09JVM optimizations for ARM
- 27:51Vector API and SIMD in Java
- 30:20Building Java apps on ARM
- 32:31Java performance ARM vs x86
- 37:27ARM cost and sustainability
- 38:54Final recommendations
related talks
40m"Cloud Cost Optimization for Java Developers"
36mWhy and How to Choose Arm64 for your Cloud Applications
34mARM Migration Made Practical
30mApache projects, let's embrace ARM!
58mEnhancing Java Applications for the Cloud
01h 02mWhat’s new for Java in the Cloud?
01h 14mModernizing and Optimizing Applications for the Cloud
01h 02mJourney to Cloud : Leave No Java Workload Behind