[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978278#comment-15978278
]
jin xing commented on SPARK-20426:
----------------------------------
[~srowen]
Thanks a lot for quick reply :)
With *spark.shuffle.service.enabled=true* , shuffle service is running under
NodeManager. NodeManager runs OOM because OneForOneStreamManager caches too
many StreamState. Yes, we can increase NodeManager's memory. But I doubt it is
a good way resolve the root cause.
> OneForOneStreamManager occupies too much memory.
> ------------------------------------------------
>
> Key: SPARK-20426
> URL: https://issues.apache.org/jira/browse/SPARK-20426
> Project: Spark
> Issue Type: Improvement
> Components: Shuffle
> Affects Versions: 2.1.0
> Reporter: jin xing
>
> Spark jobs are running on yarn cluster in my warehouse. We enabled the
> external shuffle service(*--conf spark.shuffle.service.enabled=true*).
> Recently NodeManager runs OOM now and then. Dumping heap memory, we find that
> *OneFroOneStreamManager*'s footprint is huge. NodeManager is configured with
> 5G heap memory. While *OneForOneManager* costs 2.5G and there are 5503233
> *FileSegmentManagedBuffer*. Is there any suggestions to avoid this other than
> just keep increasing NodeManager's memory? Is it possible to stop
> *registerStream* in OneForOneStreamManager? Thus we don't need to cache so
> many metadata(i.e. StreamState).
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
