[
https://issues.apache.org/jira/browse/YARN-7494?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16344656#comment-16344656
]
Wangda Tan commented on YARN-7494:
----------------------------------
[~sunilg] , sorry for the delays in review. I just took a look at the patch and
have some comments:
This patch makes multiple nodes lookup becomes a global behavior. In my mind, a
user should be able to turn this on as a per-app behavior, some apps can try to
use this feature without affecting other applications in the cluster. And then
turn it on for more apps. I'm not sure if this is a valid requirement or not.
[~cheersyang]/[~asuresh] , since you mentioned you have interests to try this
feature, could you share your thoughts here?
If we wanna to support enabling multiple nodes lookup for a single app, what we
can do is to add a method to {{CandidateNodeSet}} like
{{getSchedulerRecommendedNextNode}} and return the next SchedulerNode. With
this approach we don't need to extend CandidateNodeSet.
The abstraction of {{MultiNodeLookupPolicy}} looks good.
> Add muti node lookup support for better placement
> -------------------------------------------------
>
> Key: YARN-7494
> URL: https://issues.apache.org/jira/browse/YARN-7494
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: capacity scheduler
> Reporter: Sunil G
> Assignee: Sunil G
> Priority: Major
> Attachments: YARN-7494.001.patch, YARN-7494.v0.patch,
> YARN-7494.v1.patch
>
>
> Instead of single node, for effectiveness we can consider a multi node lookup
> based on partition to start with.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
