[jira] [Closed] (FLINK-10430) Extract scheduling-related code from Executions

["Gary Yao (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/FLINK-10430?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Gary Yao closed FLINK-10430.
----------------------------
    Resolution: Later

> Extract scheduling-related code from Executions
> -----------------------------------------------
>
>                 Key: FLINK-10430
>                 URL: https://issues.apache.org/jira/browse/FLINK-10430
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Runtime / Coordination
>    Affects Versions: 1.7.0
>            Reporter: Stefan Richter
>            Assignee: Stefan Richter
>            Priority: Major
>
> About half of the scheduling work is currently happening inside Execution objects. In particular, this is where resource profiles are constructed and the interaction with the SlotPool happens. In a first step, we should move this code out of the Executions and into a new higher level object, probably (for now) a collaborator of the ExecutionGraph. This new object becomes the first iteration of a Scheduler and the Executions are moving more into the direction of a plain data structure.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)