robobario opened a new pull request, #13482: URL: https://github.com/apache/kafka/pull/13482
Why: Using java.util.Random to generate every byte sent from the ProducerPerformance appears to be a limiting factor. Throughput of the ProducerPerformance script is higher with a file of example records as compared to randomly generated records with `--record-size`. On my machine a single thread can generate ~100MB/second of uppercase letters using java.util.Random and ~300MB/sec using java.util.SplittableRandom. This is a limit on throughput. Note: you can optimise further by expanding it from 26 letters to 32 letter generated as it is more efficient to generate a nicely distributed int when the bound is a power of two. ### Committer Checklist (excluded from commit message) - [ ] Verify design and implementation - [ ] Verify test coverage and CI build status - [ ] Verify documentation (including upgrade notes) -- 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. To unsubscribe, e-mail: jira-unsubscr...@kafka.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
