ChenMichael edited a comment on pull request #34684: URL: https://github.com/apache/spark/pull/34684#issuecomment-976849576
In order for this problem to manifest, we have to do join planning in between the time an InMemoryRelation is converted to a RDD and the time where the job executing this RDD completes. In AQE, since it repeatedly does replanning, this can happen when the InMemoryRelations are on different levels. With AQE off, it only does join planning once, so there's no scenario where part of the query materializes the InMemoryRelation and then join planning happens on another part of the query with inaccurate stats. I guess this could happen with AQE off if there are concurrent jobs sharing the same InMemoryRelation though. -- 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: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
