hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Wangda Tan (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-5889) Improve user-limit calculation in capacity scheduler
Date Wed, 16 Nov 2016 19:41:58 GMT

    [ https://issues.apache.org/jira/browse/YARN-5889?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15671396#comment-15671396

Wangda Tan commented on YARN-5889:

Thanks [~sunilg],

Haven't look at all the details, a little bit background for why this is required: This is
majorly for intra-queue user-limit preemption, today's user-limit is designed for allocation
instead of preemption. Because it only calculate a same limit for all the users in the queue,
which is not correct. So we need to calculate per-queue limit (because some users may not
need more resource, and some users may need more), this value will be used to do allocation
and preemption. The goal is to make the behavior same to today's user limit as much as possible.
But it is still possible that some minor behavior change could happen, this is why we want
an option to enable it. 

+ [~eepayne].

> Improve user-limit calculation in capacity scheduler
> ----------------------------------------------------
>                 Key: YARN-5889
>                 URL: https://issues.apache.org/jira/browse/YARN-5889
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: capacity scheduler
>            Reporter: Sunil G
>            Assignee: Sunil G
>         Attachments: YARN-5889.v0.patch
> Currently user-limit is computed during every heartbeat allocation cycle with a write
lock. To improve performance, this tickets is focussing on moving user-limit calculation out
of heartbeat allocation flow.

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