[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16447935#comment-16447935
]
ice bai commented on SPARK-20426:
-
refer with the following issue:
https://issues.apache
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15981124#comment-15981124
]
jin xing commented on SPARK-20426:
--
[~jerryshao]
I think lazy initialization can resolv
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15981117#comment-15981117
]
Apache Spark commented on SPARK-20426:
--
User 'jinxing64' has created a pull request
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978431#comment-15978431
]
jin xing commented on SPARK-20426:
--
Yes, the applications are requesting too many shuffl
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978415#comment-15978415
]
Saisai Shao commented on SPARK-20426:
-
Currently I don't have a clear fix about this
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978411#comment-15978411
]
Saisai Shao commented on SPARK-20426:
-
My simple doubt is that your large cluster / a
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978409#comment-15978409
]
jin xing commented on SPARK-20426:
--
[~jerryshao]
{quote}
Brain storm: The problem here i
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978398#comment-15978398
]
jin xing commented on SPARK-20426:
--
[~jerryshao]
Thanks a lot for looking into this jira
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978394#comment-15978394
]
jin xing commented on SPARK-20426:
--
Currently in the code, shuffle-read process is like
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978391#comment-15978391
]
Saisai Shao commented on SPARK-20426:
-
Is your yarn cluster a large cluster, and ther
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978345#comment-15978345
]
jin xing commented on SPARK-20426:
--
That's inside NodeManager(not application memory). W
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978317#comment-15978317
]
Sean Owen commented on SPARK-20426:
---
That's not in the NodeManager, right? isn't that a
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978314#comment-15978314
]
jin xing commented on SPARK-20426:
--
I posted 2 screenshots. External shuffle service of
[
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 *spa
[
https://issues.apache.org/jira/browse/SPARK-20426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978266#comment-15978266
]
Sean Owen commented on SPARK-20426:
---
The NodeManager is a YARN process. This doesn't so
15 matches
Mail list logo
