leanken commented on pull request #29304: URL: https://github.com/apache/spark/pull/29304#issuecomment-667702853
> > @leanken ... let me chew on your response. I will get back to you. > > In the meanwhile I am wondering if you can rerun the performance of the original TPCH q16 (single key not-in) with this multi-key PR and without it. I have a feeling we might have regressed the performance of the (much more common) single key case and it may make sense to retain that as a special case. Can you double check that please. > > Sure, I will do Q16 Test again. Though i have strong confident that it will not cause perf regression, because in single key not in case, it will not cause any data expansion. because 2^1 - 1 = 1 ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
