hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Omkar Vinit Joshi (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-415) Capture memory utilization at the app-level for chargeback
Date Fri, 20 Sep 2013 20:00:54 GMT

    [ https://issues.apache.org/jira/browse/YARN-415?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13773349#comment-13773349
] 

Omkar Vinit Joshi commented on YARN-415:
----------------------------------------

I am not sure if it affects this or not but today say if client asks for 2000 we round it
off.. and this patch will report it based on round off value..Is it right?
Also few questions about the scenarios
* What if container is allocated by RM and before AM launches it NM reboots (invaliding the
earlier container... so should the user be charged?) what is AM reports back saying
** It doesn't want the container - I think if should still be charged as it earlier asked
for one... but I remember there being an issue where RM allocated one more or one less than
what user requested (obviously RM scheduler issue..)
** It doesn't want the container may be because it is not able to launch it on NM for whatever
reason.
* for short running containers there is a gap between container actually finishes and is reported
back to RM are we accounting for this? Similarly AM receiving the container and RM either
allocating or processing ACQUIRED event.
* Also today container finish time is computed on RM side. I think it should be done on NM
side and communicated back to RM (implicit assumption both servers are in sync).
                
> Capture memory utilization at the app-level for chargeback
> ----------------------------------------------------------
>
>                 Key: YARN-415
>                 URL: https://issues.apache.org/jira/browse/YARN-415
>             Project: Hadoop YARN
>          Issue Type: New Feature
>          Components: resourcemanager
>    Affects Versions: 0.23.6
>            Reporter: Kendall Thrapp
>            Assignee: Andrey Klochkov
>         Attachments: YARN-415--n2.patch, YARN-415--n3.patch, YARN-415.patch
>
>
> For the purpose of chargeback, I'd like to be able to compute the cost of an
> application in terms of cluster resource usage.  To start out, I'd like to get the memory
utilization of an application.  The unit should be MB-seconds or something similar and, from
a chargeback perspective, the memory amount should be the memory reserved for the application,
as even if the app didn't use all that memory, no one else was able to use it.
> (reserved ram for container 1 * lifetime of container 1) + (reserved ram for
> container 2 * lifetime of container 2) + ... + (reserved ram for container n * lifetime
of container n)
> It'd be nice to have this at the app level instead of the job level because:
> 1. We'd still be able to get memory usage for jobs that crashed (and wouldn't appear
on the job history server).
> 2. We'd be able to get memory usage for future non-MR jobs (e.g. Storm).
> This new metric should be available both through the RM UI and RM Web Services REST API.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message