Hazelcast Uses Infinispan and Apache Ignite's Own Benchmarks and Beats Them
YardStick and RadarGun are two benchmarking tools built by Hazelcast's competitors: Infinispan / JBoss Data Grid and Ignite / GridGain. Hazelcast performed better than the competitor in each benchmark.
by
Join the DZone community and get the full member experience.
Join For FreeWe should start calling Hazelcast, Hazelfast. Hazelcast 3.6 EA shipped last week, and it is very fast. Compared to what, you ask? Well, compared to everything. How do we know this? Well, our competitors made their own comparative benchmarking tools, YardStick and RadarGun. And we ran them against Hazelcast as we were developing 3.6. We were getting faster anyway thanks to our dedicated performance team, but there is nothing like competition to spur improvement.
Hazelcast Versus Infinispan/JBoss Data Grid
On RadarGun, the Infinispan tool, we are faster in all tests with the biggest margin begin 80%.
See the full benchmark results here.
Source code for the benchmark is here: See the results of the Radar-Gun Benchmark https://github.com/Danny-Hazelcast/radargun.
Hazelcast Versus Apache Ignite/GridGain
On YardStick, the GridGain tool, we are faster in all tests with the biggest margin being 90% faster.
Source code for each benchmark is on GitHub:
- yardstick-Hazelcast https://github.com/hazelcast/yardstick-hazelcast
- yardstick-Ignite: https://github.com/hazelcast/yardstick-ignite
See the full benchmark here.
Oracle Coherence
It turns out that it is a breach of Oracle’s OTN license to publish benchmarks. So we can’t comment at all on our benchmark results compared to Oracle Coherence. But, there is no such restriction on publishing a Benchmark Suite. We can send it to you and you can run it yourself.
Published at DZone with permission of Greg Luck, DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.
Comments