alamb commented on PR #2587: URL: https://github.com/apache/arrow-datafusion/pull/2587#issuecomment-1136007628
I also agree with @tustvold that in general it will be very hard to beat the performance of using the optimized, vectorized kernels in arrow-rs and in general I think we should continue to use them whenever possible. As I mentioned in https://github.com/apache/arrow-datafusion/pull/2587#issuecomment-1133954002 I think the super valuable usecase for JIT'ed Exprs is for operations that can't easily (or at all) be vectorized, namely those doing multi column comparisons during Merge / Group -- 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]
