Rich-T-kid commented on PR #11743: URL: https://github.com/apache/arrow-rs/pull/11743#issuecomment-6566805490
A quick update: the [original benchmark results](https://github.com/apache/arrow-rs/pull/11743#issuecomment-6404461309) were misleading. [The benchmark ignored the per-batch seed for dictionary inputs, so each batch generated the same dictionary contents](https://github.com/apache/arrow-rs/commit/f3b42937d3349d11467a1849ea1241c8ea119b1e). After fixing that and rerunning the earlier implementations, the [results were much less impressive](https://github.com/apache/arrow-rs/pull/11743#issuecomment-6530623333). The PR has since been replaced with a different approach and squashed to one commit; the [latest benchmark results](https://github.com/apache/arrow-rs/pull/11743#issuecomment-6566136786) show a clear improvement. @Jefffrey could you take a second look 🙇, i've since updated the pr description as well -- 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]
