hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Weiwei Yang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-7757) Refactor NodeLabelsProvider to be more generic and reusable for node attributes providers
Date Wed, 24 Jan 2018 02:36:02 GMT

    [ https://issues.apache.org/jira/browse/YARN-7757?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16336787#comment-16336787
] 

Weiwei Yang commented on YARN-7757:
-----------------------------------

Hi [~Naganarasimha]

I thought about this after the discussion we had yesterday, but I think it is better to have
separate providers. Because we cannot force user to use labels and attributes at the same
time. User needs to have flexibility to use labels only, attributes only or both. In that
case, separated provider is the better choice. And second, if we use same class to collect
attributes stuff, it can hard be a clean implementation, which makes testing/UT much more
harder. Does that make sense?

> Refactor NodeLabelsProvider to be more generic and reusable for node attributes providers
> -----------------------------------------------------------------------------------------
>
>                 Key: YARN-7757
>                 URL: https://issues.apache.org/jira/browse/YARN-7757
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: nodemanager
>            Reporter: Weiwei Yang
>            Assignee: Weiwei Yang
>            Priority: Major
>         Attachments: YARN-7757-YARN-3409.001.patch, YARN-7757-YARN-3409.002.patch, nodeLabelsProvider_refactor_class_hierarchy.pdf
>
>
> Propose to do refactor on {{NodeLabelsProvider}}, {{AbstractNodeLabelsProvider}} to be
more generic, so node attributes providers can reuse these interface/abstract classes.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
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