You are viewing a plain text version of this content. The canonical link for it is here.

Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2021/06/03 17:44:33 UTC

[GitHub] [spark] venkata91 commented on pull request #32754: [SPARK-35613][CORE][SQL] Cache commonly occurring strings in SQLMetrics, JSONProtocol and AccumulatorV2 classes

venkata91 commented on pull request #32754:
URL: https://github.com/apache/spark/pull/32754#issuecomment-854059790


   > Can you please include the impact of these changes ?
   > For the cases we observed, what was the 'savings' for the individual changes ?
   > 
   > This will help quantify the amount of garbage saved for nontrivial spark applications.
   
   Added more details on the garbage accumulated and the impact after fixing them.
   Earlier the job used to be configured with driver memory of 40+gb and still run into heavy GC but after these fixes the driver memory is brought down to 25gb and no long Full GC pauses. Now the Full GC pauses are much more streamlined.


-- 
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org