[
https://issues.apache.org/jira/browse/YARN-7494?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16418907#comment-16418907
]
Sunil G commented on YARN-7494:
-------------------------------
Thanks [~cheersyang]
For known policies like resource-usage, this looks more easy. But for each
custom policies, we then need to define a name.
like
{code:java}
<property>
<name>yarn.scheduler.capacity.multi-node-sorting.polices</name>
<value>resource-usage,container-count,my-custom</value>
</property>{code}
then each one has to expanded like other 2 config parameter as u mentioned.
another option is that if we do a closed model like
[name=org...container,timeout=1200;name=resource-based,timeout=1000] etc.
since 2nd and 3rd config is depending on name "resource-usage", we need more
validation on name etc. I think its abt choosing which config model is more
convenient. we have a queue name based model which u just suggested, other
model like we have done for acl's etc where we use [name=,timeout=].
I ll check with [~leftnoteasy] also to pitch in to select the better one.
Thoughts?
> 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-7494.002.patch,
> YARN-7494.003.patch, YARN-7494.004.patch, YARN-7494.005.patch,
> YARN-7494.v0.patch, YARN-7494.v1.patch, multi-node-designProposal.png
>
>
> 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: [email protected]
For additional commands, e-mail: [email protected]
