hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Karthik Kambatla (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-4719) Add a helper library to maintain node state and allows common queries
Date Tue, 01 Mar 2016 02:58:18 GMT

    [ https://issues.apache.org/jira/browse/YARN-4719?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15173137#comment-15173137
] 

Karthik Kambatla commented on YARN-4719:
----------------------------------------

bq. May also be useful to expose functionality in the ClusterNodeTracker to give list of nodes
in a rack, nodes that match a label expression etc. (This can possibly be another JIRA too)
Absolutely. I wanted to move all existing common functionality into this class in this JIRA,
so we can add other helper functionality in the future. 

bq. I see that you are triggering the update thread on nodeRemoval too. I understand this
might generally be useful (since the node removal might change the node ordering), but given
this is a refactoring patch, maybe address that separately ?
removeNode does a triggerUpdate today too. I just moved it a little. 

Will fix the import and the test failures here in the next iteration. 

> Add a helper library to maintain node state and allows common queries
> ---------------------------------------------------------------------
>
>                 Key: YARN-4719
>                 URL: https://issues.apache.org/jira/browse/YARN-4719
>             Project: Hadoop YARN
>          Issue Type: Improvement
>          Components: scheduler
>    Affects Versions: 2.8.0
>            Reporter: Karthik Kambatla
>            Assignee: Karthik Kambatla
>         Attachments: yarn-4719-1.patch, yarn-4719-2.patch
>
>
> The scheduler could use a helper library to maintain node state and allowing matching/sorting
queries. Several reasons for this:
> # Today, a lot of the node state management is done separately in each scheduler. Having
a single library will take us that much closer to reducing duplication among schedulers.
> # Adding a filtering/matching API would simplify node labels and locality significantly.

> # An API that returns a sorted list for a custom comparator would help YARN-1011 where
we want to sort by allocation and utilization for continuous/asynchronous and opportunistic
scheduling respectively. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message