[
https://issues.apache.org/jira/browse/FLINK-16430?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17304010#comment-17304010
]
Till Rohrmann commented on FLINK-16430:
---------------------------------------
The scheduling works on the task level (parallel instances of an operator) and
not operator level. Also we want to failover disjunct tasks individually.
That's why we have to create the pipelined regions based on the
tasks/executions and not operators.
> FLIP-119 Pipelined Region Scheduling
> ------------------------------------
>
> Key: FLINK-16430
> URL: https://issues.apache.org/jira/browse/FLINK-16430
> Project: Flink
> Issue Type: New Feature
> Components: Runtime / Coordination
> Affects Versions: 1.11.0
> Reporter: Zhu Zhu
> Assignee: Zhu Zhu
> Priority: Major
> Fix For: 1.12.0
>
>
> Pipelined region scheduling is targeting to allow batch jobs with PIPELINED
> data exchanges to run without the risk to encounter a resource deadlock.
> More details see
> [FLIP-119|https://cwiki.apache.org/confluence/display/FLINK/FLIP-119+Pipelined+Region+Scheduling]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
