yerenkow commented on PR #41067: URL: https://github.com/apache/spark/pull/41067#issuecomment-1601959338
Not really - if we have some external process that prepares/transforms data, then it's not part of the JVM memory, and its size can't be predictable transformed into fraction of that memory. Ideally, I'd propose separate controls for a) JVM xmx, b) pod request mem, c) pod limit mem But to make it simpler, configuration for limits will do for case with burst-like load. -- 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]
