Skip to content

[AURON #2236]support compile with Celeborn 0.6 and spark 4.0 #2243

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
ftong2020 wants to merge 3 commits into
base: master
Choose a base branch
from
Open

[AURON #2236]support compile with Celeborn 0.6 and spark 4.0 #2243

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
b16a775
add support for celeborn 0.6 with spark 4.0
May 7, 2026
eaf7ae8
fix incorrect condition comparison
May 7, 2026
b40ef65
Potential fix for pull request finding
ftong2020 May 7, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions spark-extension/src/main/scala/org/apache/spark/sql/auron/NativeRDD.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,10 +67,12 @@ class NativeRDD(
override def compute(split: Partition, context: TaskContext): Iterator[InternalRow] = {
val computingNativePlan = nativePlanWrapper.plan(split, context)

// SPARK-44605: Spark 4+ refines ShuffleWriteProcessor API (early execution of NativeRDD.ShuffleWrite iterator)
// Adaptation for Spark 4.x: Defer NativeRDD.ShuffleWrite execution to ShuffleWriteProcessor.write() to align with Spark 3.x logic
if (SparkVersionUtil.isSparkV40OrGreater &&
computingNativePlan.getPhysicalPlanTypeCase == PhysicalPlanNode.PhysicalPlanTypeCase.SHUFFLE_WRITER) {
// SPARK-44605: Spark 4+ refines ShuffleWriteProcessor API (early execution of native
// shuffle-writer iterators). Adaptation for Spark 4.x: defer both native shuffle-writer
// plan types (SHUFFLE_WRITER and RSS_SHUFFLE_WRITER) to ShuffleWriteProcessor.write() to
// align with Spark 3.x logic.
if (SparkVersionUtil.isSparkV40OrGreater && (computingNativePlan.getPhysicalPlanTypeCase == PhysicalPlanNode.PhysicalPlanTypeCase.SHUFFLE_WRITER
|| computingNativePlan.getPhysicalPlanTypeCase == PhysicalPlanNode.PhysicalPlanTypeCase.RSS_SHUFFLE_WRITER)) {
Iterator.empty
} else {
NativeHelper.executeNativePlan(computingNativePlan, metrics, split, Some(context))
Expand Down
2 changes: 1 addition & 1 deletion ...c/main/scala/org/apache/spark/sql/execution/auron/shuffle/AuronRssShuffleWriterBase.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,7 +78,7 @@ abstract class AuronRssShuffleWriterBase[K, V](metrics: ShuffleWriteMetricsRepor

def rssStop(success: Boolean): Option[MapStatus]

@sparkver("3.2 / 3.3 / 3.4 / 3.5")
@sparkver("3.2 / 3.3 / 3.4 / 3.5 / 4.0")
override def getPartitionLengths(): Array[Long] = rpw.getPartitionLengthMap

override def write(records: Iterator[Product2[K, V]]): Unit = {
Expand Down
2 changes: 1 addition & 1 deletion ...la/org/apache/spark/sql/execution/auron/shuffle/celeborn/AuronCelebornShuffleWriter.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,7 +63,7 @@ class AuronCelebornShuffleWriter[K, V](
celebornPartitionWriter
}

@sparkver("3.2 / 3.3 / 3.4 / 3.5")
@sparkver("3.2 / 3.3 / 3.4 / 3.5 / 4.0")
override def getPartitionLengths(): Array[Long] = {
celebornPartitionWriter.getPartitionLengthMap
}
Expand Down
Loading