[
https://issues.apache.org/jira/browse/YARN-3215?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15120605#comment-15120605
]
Wangda Tan commented on YARN-3215:
----------------------------------
[~Naganarasimha]/[~sunilg],
Thanks for sharing your thoughts, my major concern of excluding NO_LABEL by
default is. Applications could ask nothing in the first AllocateRequest, get
headroom back and decide how many resources to ask, we cannot presume
application's behavior. Telling an app: "your headroom is 0" but there're
plenty of resources looks not good enough to me.
The solution for longer term is to add by-partition headroom, it will be a
clear protocol to return what application requested.
I won't strongly against returning only requested headroom to app, but to make
it behavior compatible (previously we returns headroom even if application
doesn't ask) and code simpler, I would prefer to add NO_LABEL by default.
Thoughts?
> Respect labels in CapacityScheduler when computing headroom
> -----------------------------------------------------------
>
> Key: YARN-3215
> URL: https://issues.apache.org/jira/browse/YARN-3215
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: capacityscheduler
> Reporter: Wangda Tan
> Assignee: Naganarasimha G R
> Attachments: YARN-3215.v1.001.patch, YARN-3215.v2.001.patch,
> YARN-3215.v2.002.patch
>
>
> In existing CapacityScheduler, when computing headroom of an application, it
> will only consider "non-labeled" nodes of this application.
> But it is possible the application is asking for labeled resources, so
> headroom-by-label (like 5G resource available under node-label=red) is
> required to get better resource allocation and avoid deadlocks such as
> MAPREDUCE-5928.
> This JIRA could involve both API changes (such as adding a
> label-to-available-resource map in AllocateResponse) and also internal
> changes in CapacityScheduler.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
