zanmato1984 commented on PR #45612: URL: https://github.com/apache/arrow/pull/45612#issuecomment-2697262557
The CI failures and regressions are not related. The benchmarking in my PR description (namely `BM_HashJoinBasic_BuildParallelism`) is not ran because it's disabled by default. But from other relevant benchmark results, namely `BM_HashJoinBasic_ProbeParallelism`, we can see nice performance up when the number of threads is high, for [example](https://conbench.ursa.dev/benchmark-results/067c675a5b8270548000487099f1d9da/) with `512k` rows hash table and `12` threads: <img width="834" alt="image" src="https://github.com/user-attachments/assets/b494095f-2eba-404e-b8b4-cd3d7a588a15"; /> I'll merge this. And much appreciate the review from @kou and @pitrou ! -- 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]
