zhengruifeng commented on a change in pull request #32328:
URL: https://github.com/apache/spark/pull/32328#discussion_r621764319
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/adaptive/OptimizeSkewedJoin.scala
##########
@@ -148,7 +148,7 @@ object OptimizeSkewedJoin extends CustomShuffleReaderRule {
/*
* This method aim to optimize the skewed join with the following steps:
Review comment:
immature thoughts on this optimization:
`OptimizeSkewedJoin` try to make sizes of shuffle partitions more even.
For bhj, should we change the *optimization goal of build side* to making
sure build side fit in memory to avoid OOM or fallback to smj.
Is it possible that the build side is skewed according current criterions,
but the shuffle partitions all fit in memory?
--
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:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
