[
https://issues.apache.org/jira/browse/YARN-7900?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16367416#comment-16367416
]
Arun Suresh commented on YARN-7900:
-----------------------------------
Thanks for working on this [~botong]..
Just went thru the patch. The original {{batchedSchedulingRequests}} is of type
{{Queue<Collection<SchedulingRequest>>}} to ensure that all
{{SchedulingRequests}} added in the same {{addSchedulingRequests}} call must be
sent in the same {{allocate}} call to the RM. The Placement Processor optimizes
the placement for a batch of scheduling requests at a time. The AM is meant to
take advantage of this, by ensuring that all co-related requests are sent in
the same allocate call.
> [AMRMProxy] AMRMClientRelayer for stateful FederationInterceptor
> ----------------------------------------------------------------
>
> Key: YARN-7900
> URL: https://issues.apache.org/jira/browse/YARN-7900
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Botong Huang
> Assignee: Botong Huang
> Priority: Major
> Attachments: YARN-7900.v1.patch, YARN-7900.v2.patch
>
>
> Inside stateful FederationInterceptor (YARN-7899), we need a component
> similar to AMRMClient that remembers all pending (outstands) requests we've
> sent to YarnRM, auto re-register and do full pending resend when YarnRM fails
> over and throws ApplicationMasterNotRegisteredException back. This JIRA adds
> this component as AMRMClientRelayer.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
