zhuqi-lucas commented on PR #15380: URL: https://github.com/apache/datafusion/pull/15380#issuecomment-2812539191
> > > > Tested latest PR, it seems regression again for current code... > > > > > > > > > Hmm... could it be because of the increased threshold? I think if we're strictly faster and keep the same heuristic we shouldn't have regressions? > > > > > > I did not change the threshold, i found it's not stable for the result, but it's will 100% improve for the sort by one column case. But for mulit column sorting, it will not improve a lot from current code. > > I meant > > > pub sort_in_place_threshold_bytes: usize, default = 4 * 1024 * 1024 > > seems to be changed? https://github.com/apache/datafusion/pull/15380#issuecomment-2809956328 Above one, this also tuning to use 4 * 1024 * 1024, so i set it to 4 * 1024 * 1024, if i change to original value, the performance 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. To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
