siddharthteotia commented on PR #18638:
URL: https://github.com/apache/pinot/pull/18638#issuecomment-4667887925
I see that with this PR we are able to avoid the GenericRow materialization
that happens today in the row major reader. However, looks like we are not
taking full benefit of this approach by eliminating the boxing overhead as
well.
The current implementation integrates collector and reader with the Object
based calls
We should instead try to do something like
```
IntVector.get(i) → int (NO BOX)
→ collect(int) / indexOfSV(int) → putInt(int)
```
Have we thought about this ? Did we intentionally leave this out of the
scope of this PR ?
--
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]
