mzabaluev commented on PR #23730: URL: https://github.com/apache/datafusion/pull/23730#issuecomment-5044745256
As seen in the `take_n_offsets_large` benchmark cases I have added in [f80450a](https://github.com/apache/datafusion/pull/23730/commits/f80450a3f5b85f71a4b63c420ba967a12fa023a7), even the `n == offsets.len() / 2 - 1` case would in fact have performed better on the split-off branch. The CPU performance optimum (ignoring memory pressure issues) looks to be somewhere below the half-point. -- 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]
