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-5773) RM recovery too slow due to LeafQueue#activateApplication()
Date Mon, 24 Oct 2016 20:47:58 GMT

    [ https://issues.apache.org/jira/browse/YARN-5773?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15603166#comment-15603166
] 

Wangda Tan commented on YARN-5773:
----------------------------------

I feel we may need a overhaul to existing activateApplication:

If we describe what activateApplications target to solve:
A set of pending applications in a queue, each application belongs to one user, different
application has different AM request, each user has a quota, and queue has a total quota,
get which application will be activated.

There's an additional questions:
If a given app's AM resource amount > AM headroom, should we skip the AM and activate following
app which AM resource amount <= AM headroom? 

If answer to the above question yes, we can maintain a map: Map<UserName, List<Application>>,
when doing application activation, we don't need to check all the apps, instead we only need
to check each user once in most cases.

> RM recovery too slow due to LeafQueue#activateApplication()
> -----------------------------------------------------------
>
>                 Key: YARN-5773
>                 URL: https://issues.apache.org/jira/browse/YARN-5773
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Bibin A Chundatt
>            Assignee: Bibin A Chundatt
>            Priority: Critical
>         Attachments: YARN-5773.0001.patch, YARN-5773.0002.patch
>
>
> # Submit application 10K application to default queue.
> # All applications are in accepted state
> # Now restart resourcemanager
> For each application recovery {{LeafQueue#activateApplications()}} is invoked.Resulting
in AM limit check to be done even before Node managers are getting registered.
> Total iteration for N application is about {{N(N+1)/2}} for {{10K}} application   {{50000000}}
iterations causing time take for Rm to be active more than 10 min.
> Since NM resources are not yet added to during recovery we should skip {{activateApplicaiton()}}




--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
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