viirya commented on code in PR #233:
URL:
https://github.com/apache/arrow-datafusion-comet/pull/233#discussion_r1540229669
##########
spark/src/main/scala/org/apache/spark/sql/comet/execution/shuffle/CometShuffleExchangeExec.scala:
##########
@@ -208,6 +216,50 @@ object CometShuffleExchangeExec extends
ShimCometShuffleExchangeExec {
dependency
}
+ /**
+ * This is copied from Spark
`ShuffleExchangeExec.needToCopyObjectsBeforeShuffle`. The only
+ * difference is that we use `BosonShuffleManager` instead of
`SortShuffleManager`.
+ */
+ private def needToCopyObjectsBeforeShuffle(partitioner: Partitioner):
Boolean = {
Review Comment:
Spark `needToCopyObjectsBeforeShuffle` will return `true` if it finds the
shuffle manager is not Spark `SortShuffleManager`. So it incurs additional row
copying for all cases which are redundant for Comet and increase memory usage.
--
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]
