avantgardnerio commented on PR #7192: URL: https://github.com/apache/arrow-datafusion/pull/7192#issuecomment-1688687429
> Where are we w.r.t. @alamb's TopK PR vs. this one? @ozankabak echoing what @alamb said, I think his PR (#7250 ) adds a general `TopK` operator which works on any incoming rows (including the output of an `AggregateExec`). This is useful and necessary for the general case, but it would still require the `AggregateExec` to hold all accumulators in memory (unless the input was sorted first). This PR is an optimization that works only for the min/max case, doesn't need sorted data, and attempts to use as little memory as possible while matching the performance of our more general purpose accumulators. -- 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]
