theirix commented on PR #21383: URL: https://github.com/apache/datafusion/pull/21383#issuecomment-4217442302
> Can you please make a separate PR to update the benchmarks? >> I can drop this evaluate_accumulator_bench benchmark to avoid noise. The focus of benchmarking should be on heavier methods like update_batch, merge_batch > I think that would be smart I've made a PR (linked) with these improvements - could you please have a look? Still, update_batch / merge_batch are not reliable enough - I've put some observations into that PR. > Also, I wonder if we can increase the time per iteration of the other benchmarks (eg. run several batches through them) to even out the noise too? Also implemented via `iter_custom` - less concise code, but with a better Criterion control -- 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]
