[
https://issues.apache.org/jira/browse/TEZ-1039?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14004045#comment-14004045
]
Bikas Saha commented on TEZ-1039:
---------------------------------
Thats correct. This checks that its affinitized container is available for
matching and so makes it waits for the container loop to bring that container
up for matching. This adds a delay that may be unnecessary but thats the best
we can do now.
Yes. firstMatch is assigned only when the request cannot match with affinity 1)
no affinity 2) affinity is not available
Will add comments.
Thanks!
> Add Container locality to TaskScheduler
> ---------------------------------------
>
> Key: TEZ-1039
> URL: https://issues.apache.org/jira/browse/TEZ-1039
> Project: Apache Tez
> Issue Type: Improvement
> Reporter: Bikas Saha
> Assignee: Bikas Saha
> Attachments: TEZ-1039.1.patch, TEZ-1039.2.patch, TEZ-1039.3.patch
>
>
> Currently, the task scheduler support node and rack locality. Proposal is to
> add another level - container locality (given a container) which means first
> try to assign to the container, then fall back to the container's node, rack
> and so on.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
