dongjoon-hyun commented on code in PR #39633:
URL: https://github.com/apache/spark/pull/39633#discussion_r1096086254
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala:
##########
@@ -1432,6 +1432,19 @@ object SQLConf {
.booleanConf
.createWithDefault(false)
+ val V2_BUCKETING_PARTIALLY_CLUSTERED_DISTRIBUTION_ENABLED =
+
buildConf("spark.sql.sources.v2.bucketing.partiallyClusteredDistribution.enabled")
+ .doc("During a storage-partitioned join, whether to allow input
partitions to be " +
+ "partially clustered, when both sides of the join are of
KeyGroupedPartitioning. At " +
+ "planning time, Spark will pick the side with less data size based on
table " +
+ "statistics, group and replicate them to match the other side. This is
an optimization " +
+ "on skew join and can help to reduce data skewness when certain
partitions are assigned " +
+ s"large amount of data. This config requires both
${V2_BUCKETING_ENABLED.key} and " +
+ s"${V2_BUCKETING_PUSH_PART_VALUES_ENABLED.key} to be enabled")
+ .version("3.4.0")
Review Comment:
If you don't mind, we still want to have this in 3.4.0, @cloud-fan and
@xinrong-meng . Please let us know your opinions.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
