Benchmark testing shows significantly better performance for OpenJDK™ 8 when using Eclipse OpenJ9 instead of Hotspot
Application performance can be measured using many different metrics, including startup time, ramp-up time, footprint, and response time, as well as throughput. At Eclipse OpenJ9, we keep a watchful eye on all of these metrics, making sensible tradeoffs and providing tuning options that allow the virtual machine (VM) to be optimized for different workloads.
The result is that OpenJDK 8 with OpenJ9 demonstrates significantly better performance than with Hotspot. Take a look at the following results:
66% smaller footprint after startup
OpenJ9 is highly optimized for cloud workloads, where minimising memory footprint is important. Even with other optimizations enabled, the footprint stays about the same.
63% smaller footprint during ramp up
Memory footprint increases rapidly when load is applied, but at steady state, OpenJDK 8 with OpenJ9 used around 63% less physical memory than OpenJDK 8 with HotSpot.
42% faster startup time
Shared classes and Ahead-of-Time (AOT) technologies typically reduce startup time. By using -Xquickstart mode as well, you can reduce startup time by up to 42%.
Although both OpenJDK 8 with OpenJ9 and OpenJDK 8 with Hotspot reach a similar peak throughput, OpenJDK 8 with OpenJ9 reaches that peak about 1 minute faster.
Faster ramp-up time in the cloud
OpenJ9 reaches peak throughput on a single CPU core in 8.5 minutes compared with 30 minutes for Hotspot. Doing more work more quickly is important for short-lived VMs running in resource-constrained environments like the cloud.
These results show that OpenJ9 achieves a good balance between (often conflicting) performance metrics: it offers excellent footprint savings and faster start-up time with the help of AOT technology, while also delivering throughput performance that is competitive with Hotspot.
Due to its low memory footprint, OpenJ9 is particularly well suited for cloud computing environments where memory savings translate into cost savings for cloud users and providers alike.
You can read the full account of how the data was generated and details of the results obtained at OpenJDK 8 performance with OpenJ9.