Konstantinos Karanasos commented on YARN-2882:

bq. I think it should be handled by some configurable/pluggable policy in NM to 
handle such extra containers in different behavior.
I am worried that this way we are restricting the capabilities of the system.
I can see applications that want to allow overcommitment but no queuing, and 
possible others that want queuing but no overcommitment.
By having parameters for queuing and overcommitment at each NM while being 
agnostic of the application, we would restrict such scenarios.

> Add ExecutionType to denote if a container execution is GUARANTEED or 
> -------------------------------------------------------------------------------
>                 Key: YARN-2882
>                 URL: https://issues.apache.org/jira/browse/YARN-2882
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: nodemanager, resourcemanager
>            Reporter: Konstantinos Karanasos
>            Assignee: Konstantinos Karanasos
>         Attachments: YARN-2882-yarn-2877.001.patch, 
> YARN-2882-yarn-2877.002.patch, YARN-2882-yarn-2877.003.patch, yarn-2882.patch
> This JIRA introduces the notion of container types.
> We propose two initial types of containers: guaranteed-start and queueable 
> containers.
> Guaranteed-start are the existing containers, which are allocated by the 
> central RM and are instantaneously started, once allocated.
> Queueable is a new type of container, which allows containers to be queued in 
> the NM, thus their execution may be arbitrarily delayed.

This message was sent by Atlassian JIRA

Reply via email to