ZhangHuiGui commented on PR #40998: URL: https://github.com/apache/arrow/pull/40998#issuecomment-2059234490
> I'll add more tests on it and push the benchmark in another pr. There is a bug when set `are_cols_in_encoding_order` to true in grouper, in order to achieve the memory alignment required when SIMD instructions access data, the columns need to be sorted. When i fix the `are_cols_in_encoding_order=false` 's bug locally, i find the performance almost same and the improvement from column resorted seems to be limited. Details: https://github.com/apache/arrow/pull/41036#discussion_r1567113203 -- 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]
