jiang13021 commented on code in PR #2609:
URL: https://github.com/apache/celeborn/pull/2609#discussion_r1680648900
##########
client-spark/spark-2/src/main/java/org/apache/spark/shuffle/celeborn/SparkUtils.java:
##########
@@ -136,6 +136,11 @@ public static int celebornShuffleId(
}
}
+ public static int getMapAttemptNumber(TaskContext context) {
+ assert (context.stageAttemptNumber() < (1 << 15) &&
context.attemptNumber() < (1 << 16));
+ return (context.stageAttemptNumber() << 16) | context.attemptNumber();
+ }
+
Review Comment:
IMO, we can specifically consider the situation involving barrier stages,
as it is safe to do so. However, considering that Spark is an extensible
open-source computing engine, deciding whether to reuse the shuffle ID based on
appShuffle's numAvailableOutputs is also a reasonable and sustainable approach.
WDYT @mridulm @waitinfuture @ErikFang @RexXiong
--
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]
