alamb commented on PR #17592: URL: https://github.com/apache/datafusion/pull/17592#issuecomment-3298529342
> @alamb Are there any benchmarks that you can run that use that flow? > > I only see the following but it benchmark bytes view which is the only non optimized case: https://github.com/apache/datafusion/blob/49d49fd92dddf55bfb22787fea17dda1a698dc4d/datafusion/physical-plan/benches/aggregate_vectorized.rs > > I see that the original PR that added it run clickbench but I don't know if clickbench have a lot of unique groups: > > * [Avoid RowConverter for multi column grouping (10% faster clickbench queries) #12269](https://github.com/apache/datafusion/pull/12269) Several of the ClickBench queries have many distinct groups so hopefully that will cover it. I kicked off the run and will check back shortly. Thank you @rluvaton -- this PR sounds quite cool. -- 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
