andygrove commented on PR #3562: URL: https://github.com/apache/datafusion-comet/pull/3562#issuecomment-3952365164
> > Thanks @andygrove one question though. One day we investigated with @mbutrovich efficiency of pregenerated queries and that time we got 40% of TPCDS queries returning no results which might affect benchmarks. We managed to improve the set to have on 18% of such queries. > > For this TPC* set how many of them return 0 rows? > > > Thanks @andygrove one question though. One day we investigated with @mbutrovich efficiency of pregenerated queries and that time we got 40% of TPCDS queries returning no results which might affect benchmarks. We managed to improve the set to have on 18% of such queries. > > For this TPC* set how many of them return 0 rows? > > I don't know. The goal for this PR is just to move them from `datafusion-benchmarks` to this repo so that we can include them in docker images for docker-compose and k8s without having dependency on another repo. > > when I do the next benchmark run I will record how many rows are returned @comphead I created https://github.com/apache/datafusion-comet/pull/3582 to start recording row counts and result hashes when running benchmarks -- 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]
