hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Abhishek Modi (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-9697) Efficient allocation of Opportunistic containers.
Date Thu, 25 Jul 2019 19:54:00 GMT

    [ https://issues.apache.org/jira/browse/YARN-9697?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16893096#comment-16893096

Abhishek Modi commented on YARN-9697:

Thanks [~elgoiri]. I will provide a POC patch soon. 

> Efficient allocation of Opportunistic containers.
> -------------------------------------------------
>                 Key: YARN-9697
>                 URL: https://issues.apache.org/jira/browse/YARN-9697
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>            Reporter: Abhishek Modi
>            Assignee: Abhishek Modi
>            Priority: Major
>         Attachments: YARN-9697.ut.patch, YARN-9697.ut2.patch
> In the current implementation, opportunistic containers are allocated based on the
number of queued opportunistic container information received in node heartbeat. This
information becomes stale as soon as more opportunistic containers are allocated on that
> Allocation of opportunistic containers happens on the same heartbeat in which AM asks
for the containers. When multiple applications request for Opportunistic containers,
containers might get allocated on the same set of nodes as already allocated containers on
the node are not considered while serving requests from different applications. This can
lead to uneven allocation of Opportunistic containers across the cluster leading to increased
queuing time 

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