tschwarzinger commented on PR #25328: URL: https://github.com/apache/datafusion/pull/25328#issuecomment-5886105559
@athlcode thank you for working on this! Sorry for the late reply, I've been on vacation. The approach looks sound by quickly scrolling over it. I do have two reservations before doing a full review: 1. Did you talk to some of the other people whether this is something they want in DataFusion? That was one of my main question when creating the issue. I've added a comment to https://github.com/apache/datafusion/issues/25610 to maybe get some further opinions on it. 2. Do you have a benchmark that shows the performance benefits of this approach? For this to work, we would need a query where the more accurate statistics lead to a different join order and thus better performance. Relevant issues are: - https://github.com/apache/datafusion/issues/8227 - https://github.com/apache/datafusion/issues/25610 - https://github.com/apache/datafusion/issues/15885 -- 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]
