srowen commented on issue #22015: 
[SPARK-20286][SPARK-24786][Core][DynamicAllocation] Release executors on 
unpersisting RDD
URL: https://github.com/apache/spark/pull/22015#issuecomment-464494103
 
 
   I haven't thought this through much, but, we don't know whether the cached 
data will be reused by the very next task. It might be aggressive to remove the 
executor with cached data just because it has not tasks at the moment. 
   
   I get that it can be a problem, if user codes isn't cleaning up cached data, 
as it may take some time until the RDD is GCed (if at all). I think the current 
design choice considers that an issue for users to clean up, because the 
alternative also creates issues.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to