[GitHub] [spark] cloud-fan commented on a change in pull request #31113: [SPARK-34061][SQL] DISTINCT the INTERSECT children

[GitBox <gi...@apache.org>]

cloud-fan commented on a change in pull request #31113:
URL: https://github.com/apache/spark/pull/31113#discussion_r589321776



##########
File path: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
##########
@@ -1663,8 +1663,12 @@ object ReplaceDeduplicateWithAggregate extends Rule[LogicalPlan] {
 /**
  * Replaces logical [[Intersect]] operator with a left-semi [[Join]] operator.
  * {{{
- *   SELECT a1, a2 FROM Tab1 INTERSECT SELECT b1, b2 FROM Tab2
- *   ==>  SELECT DISTINCT a1, a2 FROM Tab1 LEFT SEMI JOIN Tab2 ON a1<=>b1 AND a2<=>b2

Review comment:
       This seems like we push down the aggregate operator through the join operator. It's usually beneficial, but can cause perf regression if the join children do not have many duplications and the aggregate operator can't reduce the data volume.




----------------------------------------------------------------
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:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org