[ 
https://issues.apache.org/jira/browse/FLINK-12229?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16833241#comment-16833241
 ] 

ryantaocer edited comment on FLINK-12229 at 5/5/19 6:40 AM:
------------------------------------------------------------

Maybe we could introduce a new trigger {{PROGRESS}} the percentage of which the 
input result partitions have finished to schedule each downstream consumer 
task. This {{PROGRESS}} parameter is of range 0~1, and 0 corrresponds to 
current {{ANY}} and 1 corresponds to current {{ALL}} respectively. In practice, 
it can be configured to about 1 saying 0.8 to balance the speedup advantage and 
possible resource deadlock.


was (Author: ryantaocer):
Maybe we could introduce a new scheduling trigger {{PROGRESS}} the percentage 
of which the input result partitions have finished to schedule each downstream 
consumer task. This {{PROGRESS}} parameter is of range 0~1, and 0 corrresponds 
to current {{ANY}} and 1 corresponds to current {{ALL}} respectively. In 
practice, it can be configured to about 1 saying 0.8 to balance the speedup 
advantage and possible resource deadlock.

> Implement Lazy Scheduling Strategy
> ----------------------------------
>
>                 Key: FLINK-12229
>                 URL: https://issues.apache.org/jira/browse/FLINK-12229
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Runtime / Coordination
>            Reporter: Gary Yao
>            Assignee: BoWang
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> Implement a {{SchedulingStrategy}} that covers the functionality of 
> {{ScheduleMode.LAZY_FROM_SOURCES}}, i.e., vertices are scheduled when all the 
> input data are available.
> Acceptance Criteria:
>  * New strategy is tested in isolation using test implementations (i.e., 
> without having to submit a job)



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

Reply via email to