Jackie-Jiang opened a new pull request, #9604:
URL: https://github.com/apache/pinot/pull/9604

   - Do not cache record reader in the segment because it might allocate too 
much memory (especially when there are compressed raw index columns, where we 
need to pre-allocate a big direct buffer)
   - Unify the code of reading record/value from segment
   - Use `PinotSegmentRecordReader` for batch read so that we get the same 
performance.
   - For partial-upsert, since we need to do point lookup record read, we will 
allocate buffers for each read, and might observe worse performance. Since it 
is not on query path, allocating per read should be okay.


-- 
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]

Reply via email to