dongjoon-hyun commented on pull request #34846: URL: https://github.com/apache/spark/pull/34846#issuecomment-1002410066
Thank you. For the customers who have many jobs, the previous values of `memoryManager.pageSizeBytes` are a kind of magic number per job because it depends on `cores` and `memory`. So, it's hard to recover their jobs one by one. I thought something like the following which disables the new logic completely for them. ```scala case Some(heapRegionSize) if isEnabled() && tungstenMemoryMode == MemoryMode.ON_HEAP => ``` -- 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]
