Dandandan commented on issue #790: URL: https://github.com/apache/arrow-datafusion/issues/790#issuecomment-894021924
 Some profiling output of the gby_null_new branch. I think it shows that: * There is quite some time spent around creating/dropping/mutating `ScalarValue` and converting it to an Array again. The same applies for the "accumulators". I think the "keep values/state in typed contiguous arrays" could mostly remove this overhead. * slicing / taking arrays takes some time and some intermediate `Vec` creation. This relates to the discussion to update states by index instead of grouping them in the aggregation. -- 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]
