hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yufei Gu (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-6516) FairScheduler:the algorithm of assignContainer is so slow for it only can assign a thousand containers per second
Date Tue, 25 Apr 2017 18:57:04 GMT

    [ https://issues.apache.org/jira/browse/YARN-6516?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15983443#comment-15983443
] 

Yufei Gu commented on YARN-6516:
--------------------------------

IIUC, there were only 40 containers in your cluster over time. More likely it is not a performance
issue. Some thoughts and questions:
# Continuous scheduling is unnecessary for a cluster with 2500 nodes. Turn on the assignmultiple
should be enough. 
# There were 39 containers for 39 apps in your cluster. I guess these containers are all for
AM. No tasks for apps? That's impossible for MR jobs. 
# How did you conclude "assigning a  container took 3ms"?

> FairScheduler:the algorithm of assignContainer is so slow for it only can assign a thousand
containers per second
> -----------------------------------------------------------------------------------------------------------------
>
>                 Key: YARN-6516
>                 URL: https://issues.apache.org/jira/browse/YARN-6516
>             Project: Hadoop YARN
>          Issue Type: Improvement
>          Components: fairscheduler
>    Affects Versions: 2.7.2
>            Reporter: JackZhou
>




--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org


Mime
View raw message