HippoBaro commented on PR #11267: URL: https://github.com/apache/arrow-rs/pull/11267#issuecomment-6089665311
Thank you all! > In my mind the problem isn't just parallel execution of the benchmarks, it is also being able to sift through the data that comes back. Maybe with LLMs now it isn't as hard to review and verify the benchmark data @alamb Agreed. I personally find the `critcmp` output pretty hard to scan for meaningful deltas. Even something as simple as sorting benchmarks by the magnitude of the relative difference would make regressions and improvements much easier to spot, IMO. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
