[
https://issues.apache.org/jira/browse/YARN-7437?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16243162#comment-16243162
]
Wangda Tan commented on YARN-7437:
----------------------------------
Thanks [~kkaranasos] for these comments,
bq. AppPlacementAllocator: comments need some fixing..
Done and filed YARN-7457.
bq. CandidateNodeSet: let's add some more comments in the beginning of the
class. Also worth mentioning we currently consider a single node.
Added more details. I would prefer to keep "single node" lookup as part of
scheduler/allocator implementation since it is not purpose of the
CandidateNodeSet class itself. I have a comment inside:
{code}
LocalityAppPlacementAllocator#getPreferredNodeIterator
{code}
And addressed all other comments except:
bq. I would prefer the variable candidates to be named something like
nodeCandidates in various classes, but that's not a big deal, I don't mind if
you leave it as is.
I think candidate should be good enough, and changing to nodeCandidates have to
reformat lots of lines.
> Give SchedulingPlacementSet to a better name.
> ---------------------------------------------
>
> Key: YARN-7437
> URL: https://issues.apache.org/jira/browse/YARN-7437
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Wangda Tan
> Assignee: Wangda Tan
> Attachments: YARN-7437.001.patch, YARN-7437.002.patch,
> YARN-7437.003.patch
>
>
> Currently, the SchedulingPlacementSet is very confusing. Here're its
> responsibilities:
> 1) Store ResourceRequests. (Or SchedulingRequest after YARN-6592).
> 2) Decide order of nodes to allocate when there're multiple node candidates.
> 3) Decide if we should reject node for given requests.
> 4) Store any states/cache can help make decision for #2/#3
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
