Dandandan commented on PR #7192:
URL: 
https://github.com/apache/arrow-datafusion/pull/7192#issuecomment-1683672115

   Looking nice 🚀 
   
   Some suggestions:
   
   * Support non-String `ID` (without converting to `String`) + add a bench for 
this case (**_this is potentially compared to normal aggregations as it causes 
to convert primitive types to `String`, force hashing as string, etc. _**)
   * Remove `SortOptions` from `HeapItem`
   * See if we can further reduce allocations - flamegraph[1] suggests a large 
part of the time in the aggregation is spent doing allocations both in `btree` 
and outside of it
   * Misc cleanups
   
   [1]
   
![flamegraph](https://github.com/apache/arrow-datafusion/assets/163737/1631fbd4-42c2-4886-a1e5-a016dddb29dc)
   
   


-- 
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]

Reply via email to