I'm looking for a better understanding on how good/ill optimized is the native Oracle JVM regarding the usage of processor/architecture specific features (eg: Cache layout, SIMD instructions, OOO Execution Pipelines, etc ... ).
Also. Which would be a good example of a processor-optimized JVM implementation ?
I would add the following to Peter's answer.
Historically, the "native" JVM (from Sun) is based on HostSpot just-in-time compilation. Alternative JVM's could be found around, such as JRockit. It was known that Hotspot JVM is better suited for UI desktop applications whereas JRockit is designed for best performance and fast executions.
Nowadays, Oracle is the owner of both Sun's JVM (HostSpot) and JRockit. The first one is shipped with the standard JRE, the latest is shipped with Oracle fusion. So, both are "native Oracle JVM's".
Note that Oracle is merging the best of the two worlds since Java 8.
An interesting discussion on both implementations can be found at http://blog.dbi-services.com/a-comparison-of-java-virtual-machines-hotspot-jvm-vs-jrockit-jvm/
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Control Theory
- Control Systems