jolshan edited a comment on pull request #9944: URL: https://github.com/apache/kafka/pull/9944#issuecomment-776097476
> > Also, it seems that for more partitions, the performance with this PR is noticeably worse than trunk? > > ``` > FetchRequestBenchmark.testSerializeFetchRequestForReplica 20 500 avgt 15 671785.374 ± 2631.210 ns/op > FetchRequestBenchmark.testSerializeFetchRequestForReplica 20 1000 avgt 15 1235326.349 ± 6101.974 ns/op > > trunk > FetchRequestBenchmark.testSerializeFetchRequestForReplica 20 500 avgt 15 1407744.600 ± 10321.934 ns/op > FetchRequestBenchmark.testSerializeFetchRequestForReplica 20 1000 avgt 15 2901845.093 ± 59557.128 ns/op > ``` Just to clarify this, the top is the Fetch branch, so I think it is better than trunk I do want to take another look at the FetcherThread and FetchSession benchmarks which are slightly worse. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org