cbalci opened a new pull request, #10209:
URL: https://github.com/apache/pinot/pull/10209

   Updated the `PinotGrpcServerDataFetcher` to return an `Iterator[DataTable]` 
instead of `List[DataTable]` which reduces memory usage on Spark reader side.
   
   **Testing**
   This is a purely performance related improvement and there are no functional 
changes, so I'm not adding specific unit tests. 
   
   Our perf tests indicated significant improvements in GC behavior on Spark 
executors, especially for very large scans (>1M rows). Here is a comparison for 
"Task time (GC time)" from sample runs.
   
   
   
   | From this |  | To this |
   |----------|----|------|
   | ![Screenshot 2023-01-31 at 3 24 35 
PM](https://user-images.githubusercontent.com/212284/215906895-7ae3b92d-f848-4df1-9459-8754b30e793f.png)
 | -> | ![Screenshot 2023-01-31 at 3 24 02 
PM](https://user-images.githubusercontent.com/212284/215906929-81f8ddbd-4435-4f2d-9624-94140752229f.png)
 |
   
   `performance`


-- 
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