[ 
https://issues.apache.org/jira/browse/YARN-7457?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16291951#comment-16291951
 ] 

Wangda Tan commented on YARN-7457:
----------------------------------

[~kkaranasos], thanks for posting the comment. I agree that missed 
opportunistic is not a good protocol. The cost function is an interesting idea, 
with that we can even consider crazy ideas like training a deep neural network 
to do the placement in the future.

I'm not sure if specifying urgency number is a good idea or not, the time-based 
delay is easy for people to understand.

In general, since we're all agree that delay scheduling logic should be 
decoupled, I propose to do it as the first step and discuss more while working 
on that.

bq. I also see the need for each application to specify how "urgent" its 
requests are. That is, one app might want to relax locality after 1 sec and 
another might want to relax after 1 min.
Agree. This is why we introduced delayed-or in YARN-6592. In addition to that, 
I think we can add a common String to String scheduling hint to 
AppSubmissionContext/RegisterAMRequest for legacy apps which using 
ResourceRequest and other tasks (such as YARN-7494) can benefit from the 
scheduling hint maps. 





> Delay scheduling should be an individual policy instead of part of scheduler 
> implementation
> -------------------------------------------------------------------------------------------
>
>                 Key: YARN-7457
>                 URL: https://issues.apache.org/jira/browse/YARN-7457
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>            Reporter: Wangda Tan
>            Assignee: Tao Yang
>
> Currently, different schedulers have slightly different delay scheduling 
> implementations. Ideally we should make delay scheduling independent from 
> scheduler implementation. Benefits of doing this:
> 1) Applications can choose which delay scheduling policy to use, it could be 
> time-based / missed-opportunistic-based or whatever new delay scheduling 
> policy supported by the cluster. Now it is global config of scheduler.
> 2) Make scheduler implementations simpler and reusable.
> h2. {color:red}Running design doc: 
> https://docs.google.com/document/d/1rY-CJPLbGk3Xj_8sxre61y2YkHJFK8oqKOshro1ZY3A/edit#heading=h.xnzvh9nn283a{color}



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to