Rich-T-kid commented on PR #26973: URL: https://github.com/apache/datafusion/pull/26973#issuecomment-6742371282
I followed up on Q28 because it was the only material regression in the current hosted ClickBench run. Four balanced local runs (`main / PR / PR / main`, five iterations each) gave a pooled median of 2127.83 ms on main and 2188.88 ms on this PR (+2.9%). The paired runs changed direction: the first pair had the PR 12.0% slower, while the reverse pair had it 14.3% faster. The earlier hosted run also classified Q28 as unchanged. I do not see a reproducible Q28 regression locally, so I have left the page-encoding policy unchanged. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
