[ 
https://issues.apache.org/jira/browse/YARN-4597?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15569508#comment-15569508
 ] 

Karthik Kambatla commented on YARN-4597:
----------------------------------------

Thanks for working on this, [~asuresh]. The approach looks reasonable to me. 
The patch is pretty big; it might be easier to use Github PR or RB for a 
thorough review - especially for minor comments.

High-level comments on the patch, keeping YARN-1011 work in mind:
# Really like that we are getting rid of QueuingContainer* classes, letting 
both guaranteed/queued containers go through the same code path 
# In ContainerScheduler, I see there are two code paths leading to starting a 
container for when enough resources are available or not. Did you consider a 
single path where we queue containers directly and let another thread launch 
them? This thread could be triggered immediately on queuing a container, on 
completion of a container, and periodically for cases where we do resource 
oversubscription.
# The methods for killing containers as needed all seem to be hardcoded to only 
consider allocated resources. Can we abstract it out further to allow for 
passing either allocation or utilization based on whether oversubscription is 
enabled. 
# Relatively minor: resourcesToFreeUp is initialized to container allocation on 
the node. Shouldn't it be initialized to zero? May be I am missing something. 

> Add SCHEDULE to NM container lifecycle
> --------------------------------------
>
>                 Key: YARN-4597
>                 URL: https://issues.apache.org/jira/browse/YARN-4597
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: nodemanager
>            Reporter: Chris Douglas
>            Assignee: Arun Suresh
>         Attachments: YARN-4597.001.patch, YARN-4597.002.patch
>
>
> Currently, the NM immediately launches containers after resource 
> localization. Several features could be more cleanly implemented if the NM 
> included a separate stage for reserving resources.



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

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

Reply via email to