gengliangwang commented on a change in pull request #34092:
URL: https://github.com/apache/spark/pull/34092#discussion_r716003560
##########
File path: core/src/main/scala/org/apache/spark/status/AppStatusListener.scala
##########
@@ -1294,6 +1330,22 @@ private[spark] class AppStatusListener(
cleanupCachedQuantiles(key)
key
}
+ }
+
+ private def cleanupStages(count: Long): Unit = {
+ val countToDelete = calculateNumberToRemove(count,
conf.get(MAX_RETAINED_STAGES))
+ if (countToDelete <= 0L) {
+ return
+ }
+
+ // SPARK-36827: For better performance and avoiding OOM, here we use a
optimized method for
+ // cleaning the StageDataWrapper and RDDOperationGraphWrapper
data if Spark is
+ // using InMemoryStore.
+ val stageIds = if (kvstore.usingInMemoryStore) {
+ cleanupStagesWithInMemoryStore(countToDelete)
Review comment:
@srowen +1. I think we can have an enhanced version of in-memory
`ElementTrackingStore`.
I was trying to merge this one in 3.2.0 so this is a temporary solution.
--
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]
