cloud-fan commented on a change in pull request #26516: [SPARK-29893] improve
the local shuffle reader performance by changing the reading task number from 1
to multi.
URL: https://github.com/apache/spark/pull/26516#discussion_r347215506
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/adaptive/OptimizeLocalShuffleReader.scala
##########
@@ -28,17 +28,19 @@ import
org.apache.spark.sql.execution.joins.{BroadcastHashJoinExec, BuildLeft, B
import org.apache.spark.sql.internal.SQLConf
object BroadcastJoinWithShuffleLeft {
- def unapply(plan: SparkPlan): Option[(QueryStageExec, BuildSide)] = plan
match {
- case join: BroadcastHashJoinExec if
ShuffleQueryStageExec.isShuffleQueryStageExec(join.left) =>
- Some((join.left.asInstanceOf[QueryStageExec], join.buildSide))
+ def unapply(plan: SparkPlan): Option[(SparkPlan, BuildSide)] = plan match {
+ case join: BroadcastHashJoinExec if join.left.isInstanceOf[QueryStageExec]
Review comment:
shall we use `ShuffleQueryStageExec.isShuffleQueryStageExec(join.left)`?
`join.left.isInstanceOf[QueryStageExec]` can be broadcast exchange.
----------------------------------------------------------------
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]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
