hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stefanlee (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-7672) hadoop-sls can not simulate huge scale of YARN
Date Sun, 25 Feb 2018 12:19:00 GMT

    [ https://issues.apache.org/jira/browse/YARN-7672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16376064#comment-16376064

stefanlee commented on YARN-7672:

[~yufeigu] [~ywskycn]  could you please tell me whether or not the value of *yarn.scheduler.fair.max.assign* can increase
the containers which the FairScheduler assigned , i set this value to *2* and run SLS in
the first round,  then next round i update this value to *8* and run SLS again . After that,
i  found there is no different between their assign  ability. (i have enabled *yarn.scheduler.fair.assignmultiple*
and disabled *continuous-scheduling*)


> hadoop-sls can not simulate huge scale of YARN
> ----------------------------------------------
>                 Key: YARN-7672
>                 URL: https://issues.apache.org/jira/browse/YARN-7672
>             Project: Hadoop YARN
>          Issue Type: Improvement
>            Reporter: zhangshilong
>            Assignee: zhangshilong
>            Priority: Major
>         Attachments: YARN-7672.patch
> Our YARN cluster scale to nearly 10 thousands nodes. We need to do scheduler pressure
> Using SLS,we start  2000+ threads to simulate NM and AM. But  cpu.load very high to 100+.
I thought that will affect  performance evaluation of scheduler. 
> So I thought to separate the scheduler from the simulator.
> I start a real RM. Then SLS will register nodes to RM,And submit apps to RM using RM

This message was sent by Atlassian JIRA

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

View raw message