-
Notifications
You must be signed in to change notification settings - Fork 58
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Improve simple map benchmark #467
Improve simple map benchmark #467
Conversation
de88269
to
4b9e3ef
Compare
I liked the new benchmark using the client to its fullest performance. Thanks for noticing the old benchmark was self throttling. About this new one, is it possible to re-introduce the parameterized put-get-remove options? We can set PUT_PERCENTAGE to 100 to consistently test put throughput. Also, I think having a key space and some randomness makes more sense because we would like to test it against a cluster of multiple instances. Having a single key makes the client use only a single instance. Even with a single instance scenario, we are limiting hazelcast member to utilize only a single partition. Therefore, we remove parallelization from member by using a single key. Also, I would like to have benchmark code compatible with node.js 4 as long as we support it. |
@mustafaiman As for the node 4, I'll downgrade syntax to support it. |
@puzpuzpuz sounds awesome |
4b9e3ef
to
c179115
Compare
Also added a readme for the benchmark
c179115
to
fb4c451
Compare
@mustafaiman |
Also added a readme for the benchmark
SimpleMapBenchmark.js
benchmark with two simple parallel ops benchmark.MapRandomOpRunner
runner is using the original random operation generator.MapPutRunner
runner removes randomness factor and dependency on server-side from the benchmark: all operations now aremap.put('foo', 'bar')
.Old benchmark showed about 16-17K ops/sec on my laptop, while the new one (namely
MapPutRunner
) shows about 50-60K ops/sec.