OpenJ9 Logo

Performance Overview

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. We regularly test and optimize OpenJ9 performance when running the most popular Java frameworks such as Open Liberty, Quarkus, SpringBoot, and Micronaut. We're proud of our results.

OpenJDK 11 performance with Eclipse OpenJ9

OpenJDK 11 with OpenJ9 significantly outperforms HotSpot on Liberty startup, ramp up, and footprint.

51% faster startup time

By using shared classes cache and AOT technology, OpenJ9 starts in roughly half the time it takes HotSpot.

50% smaller footprint after startup

After startup, the OpenJ9 footprint is half the size of HotSpot, which makes it ideal for cloud workloads.

Faster ramp-up time in the cloud

OpenJ9 reaches peak throughput much faster than HotSpot making it especially suitable for running short-lived applications.

33% smaller footprint during load

Consistent with the footprint results after startup, the OpenJ9 footprint remains much smaller than HotSpot when load is applied.

Show me performance details

With version of Open Liberty and OpenJ9, Open Liberty managed to cut the startup time to their goal of almost exactly one second.

Open Liberty blog 

Payara found that with datagrid enabled, Payara Micro with class data sharing, booted in about 6 seconds, improving startup time by almost 40%.

Payara Enterprise 

OpenJDK 11 performance with Quarkus and Eclipse OpenJ9

Using Quarkus with OpenJ9 instead of HotSpot makes Java even more supersonic and subatomic!

66% faster startup time

Quarkus might be supersonic, but running it with OpenJ9 instead of HotSpot reduces startup time by a further 66%.

18% smaller footprint after startup

The ultra-supersonic startup time comes at some tradeoff to footprint, but OpenJ9 still has an edge over HotSpot.

Faster ramp-up time in the cloud

For short-lived cloud workloads running on Quarkus, OpenJ9 delivers excellent ramp-up performance compared to HotSpot.

18% smaller footprint during load

When ramp-up completes, as well as getting there faster, OpenJ9 still retains a smaller footprint than HotSpot.

Show me performance details

Trendyol Tech found 40% faster startup time and 73 MB less memory footprint when they ran Undertow + Rest + Couchbase Client with OpenJ9.

Trendyol Tech blog 

As reported on twitter, Micronautfw startup with OpenJ9 is as fast as Node.js.

Graeme Rocher, Micronaut 

OpenJDK 8 performance with Eclipse OpenJ9

Testing shows similar results for OpenJDK 8; OpenJ9 demonstrates significantly better performance 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%.

66% smaller footprint after startup

OpenJ9 is highly optimized for cloud workloads, where minimizing memory footprint is important. Out of the box, the footprint is 66% smaller than HotSpot.

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. Ideal for short-lived VMs running in the cloud.

63% smaller footprint during load

Memory footprint increases rapidly when load is applied. However, at steady state, OpenJ9 consistently used around 63% less physical memory than HotSpot.

Show me performance details

Vert.x found that OpenJ9 was very efficient with respect to memory consumption, without compromising the latency.

Eclipse Vert.x how-to guide 

As reported on the Amis technology blog, when testing a number of JVMs, OpenJDK with OpenJ9 was the fastest to start for every framework.

AMIS technology blog 
Eclipse Incubator Logo
OpenJ9 is an Eclipse Incubator Project
OpenJ9 is an Eclipse Incubator Project
Eclipse Incubator Logo