You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Eron Wright (JIRA)" <ji...@apache.org> on 2017/02/28 23:47:45 UTC

[jira] [Commented] (FLINK-4427) Implement container releasing logic (Standalone / Yarn / Mesos)

    [ https://issues.apache.org/jira/browse/FLINK-4427?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15889134#comment-15889134 ] 

Eron Wright  commented on FLINK-4427:
-------------------------------------

Would it be possible to structure the releasing logic to support numerous strategies?   The default strategy might be to release an idle task manager after some time.   Another strategy might be to maintain a small pool of TMs (to support short interactive-style jobs) while bursting to larger numbers as needed.   Maybe a pool per  resource profile.

> Implement container releasing logic (Standalone / Yarn / Mesos)
> ---------------------------------------------------------------
>
>                 Key: FLINK-4427
>                 URL: https://issues.apache.org/jira/browse/FLINK-4427
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Cluster Management
>            Reporter: Kurt Young
>
> Currently we only have allocation logic for SlotManager / ResourceManager, for some batch job, slots that already finished can be released, thus should trigger container release in different cluster modes.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)