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

Posted to issues@flink.apache.org by "Till Rohrmann (Jira)" <ji...@apache.org> on 2019/09/13 13:03:00 UTC

[jira] [Comment Edited] (FLINK-5136) Aggregation of slot ResourceProfile and framework memory by ResourceManager

    [ https://issues.apache.org/jira/browse/FLINK-5136?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16929180#comment-16929180 ] 

Till Rohrmann edited comment on FLINK-5136 at 9/13/19 1:02 PM:
---------------------------------------------------------------

Superseded by [FLIP-53|https://cwiki.apache.org/confluence/display/FLINK/FLIP-53%3A+Fine+Grained+Operator+Resource+Management]


was (Author: till.rohrmann):
Superseded by [FLIP-53|https://cwiki.apache.org/confluence/display/FLINK/FLIP-53%3A+Fine+Grained+Resource+Management]

> Aggregation of slot ResourceProfile and framework memory by ResourceManager
> ---------------------------------------------------------------------------
>
>                 Key: FLINK-5136
>                 URL: https://issues.apache.org/jira/browse/FLINK-5136
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Runtime / Coordination
>            Reporter: zhijiang
>            Assignee: zhijiang
>            Priority: Major
>
> When *ResourceManager* requests container resource from cluster, the framework memory should be considered together with slot resource.
> Currently the framework memory is mainly used by *MemoryManager* and *NetworkBufferPool* in *TaskManager*, and this memory can be got from configuration.



--
This message was sent by Atlassian Jira
(v8.3.2#803003)