cloud-fan commented on a change in pull request #33123:
URL: https://github.com/apache/spark/pull/33123#discussion_r660368966
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/adaptive/ShufflePartitionsUtil.scala
##########
@@ -173,7 +176,11 @@ object ShufflePartitionsUtil extends Logging {
if (numPartitions > start) {
val partitionSpecs = coalescePartitions(
partitionIndices(start), partitionIndices.last + 1, validMetrics,
targetSize)
- newPartitionSpecsSeq.zip(partitionSpecs).foreach(spec => spec._1 ++=
spec._2)
+ assert(numPartitions > 0)
+ // skip empty partition iff all shuffle are empty after coalesced
+ if (!partitionSpecs.forall(specs => specs.size == 1 &&
specs.head.dataSize.contains(0))) {
Review comment:
I looked into `coalescePartitions`, it does skip empty partitions, but
it needs to at least return one `CoalescedPartitionSpec`.
I think we can pass one more boolean flag to `coalescePartitions` to allow
it to return no `CoalescedPartitionSpec`, for coalescing partitions between
skewed partitions.
--
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: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
