SteNicholas commented on code in PR #3601:
URL: https://github.com/apache/celeborn/pull/3601#discussion_r2852122171
##########
common/src/main/scala/org/apache/celeborn/common/CelebornConf.scala:
##########
@@ -6854,4 +6858,19 @@ object CelebornConf extends Logging {
.booleanConf
.createWithDefault(false)
+ val SHUFFLE_WRITE_LIMIT_ENABLED: ConfigEntry[Boolean] =
+ buildConf("celeborn.client.spark.shuffle.write.limit.enabled")
+ .categories("client")
+ .doc("Enable shuffle write limit check to prevent cluster resource
exhaustion.")
+ .version("0.7.0")
+ .booleanConf
+ .createWithDefault(false)
+
+ val SHUFFLE_WRITE_LIMIT_THRESHOLD: ConfigEntry[Long] =
+ buildConf("celeborn.client.spark.shuffle.write.limit.threshold")
+ .categories("client")
+ .doc("Shuffle write limit threshold, exceed to cancel oversized
shuffle.")
+ .version("0.7.0")
+ .bytesConf(ByteUnit.BYTE)
+ .createWithDefaultString("5TB")
Review Comment:
IMO, `celeborn.client.spark.shuffle.write.limit.threshold` is better to be
optional. The default value of threshold should be common for production
environment.
--
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]
