hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Panagiotis Garefalakis (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-7612) Add Placement Processor and planner framework
Date Tue, 12 Dec 2017 14:56:00 GMT

    [ https://issues.apache.org/jira/browse/YARN-7612?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16287710#comment-16287710
] 

Panagiotis Garefalakis commented on YARN-7612:
----------------------------------------------

Thanks [~asuresh] for the patch and [~leftnoteasy] for the input.
Some comments from my side:
* I like the latest common interfaces split - not sure why PlacementAlgorithm abstract class
is not part of  constraint.spi package as well?
* RejectedReason ENUMS look confusing to me - something like  INVALID_PLACEMENT_REJECTION
and INFEASIBLE_PLACEMENT_REJECTION seem more reasonable to me.
* Currently PlacementConstraintsManager interface considers constraints as part of appIDs
but we also need to support **cluster wide* constraints - so I would add another more generic
setter and getter as well 
* Looking at the SamplePlacementAlgorithm implementation one thing we are not currently taking
into to account is the previously placed SchedulingRequests tags of the current Batch (by
previously I mean in the previous iterations) where the containers are still not launched
thus the TagManager is not aware of their existence. One way to solve this would be having
the algorithm implementation to keep an extra data structure with the placed tags, another
would be to extend the TagManager to keep a temporary mapping. Any thoughts on that?

Minor:
* Unused imports interface ResourceScheduler
* Typo (placementConstriantsManager) affecting RMContext and RMActiveServiceContext
* AbstractYarnScheduler changes do not seem to apply cleanly to me


> Add Placement Processor and planner framework
> ---------------------------------------------
>
>                 Key: YARN-7612
>                 URL: https://issues.apache.org/jira/browse/YARN-7612
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>            Reporter: Arun Suresh
>            Assignee: Arun Suresh
>         Attachments: YARN-7612-YARN-6592.001.patch, YARN-7612-YARN-6592.002.patch, YARN-7612-YARN-6592.003.patch,
YARN-7612-YARN-6592.004.patch, YARN-7612-YARN-6592.005.patch, YARN-7612-v2.wip.patch, YARN-7612.wip.patch
>
>
> This introduces a Placement Processor and a Planning algorithm framework to handle placement
constraints and scheduling requests from an app and places them on nodes.
> The actual planning algorithm(s) will be handled in a YARN-7613.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org


Mime
View raw message