[ 
https://issues.apache.org/jira/browse/FLINK-17295?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17539914#comment-17539914
 ] 

Zhu Zhu edited comment on FLINK-17295 at 5/20/22 4:31 AM:
----------------------------------------------------------

I was not concerning about it because PartitionRequests are sent and processed 
by TMs in a distributed way, unlike TDD shipping which relies on the master 
node.

To further eliminating concerns, I did an experiment to see the job execution 
time with/without the proposed change.

The testing job is a word count job of no data, parallelism=1000, run 3 times 
for each case. The measured execution time is from deploying started to job 
finishes (to exclude container allocation time to stabilize the result).

The average execution time with the change is 7.4s (7.6,7.5,7.0). The average 
execution time without the change is 7.9s(8.7,6.9,8.0). Therefore, I think the 
change is acceptable.


was (Author: zhuzh):
I was not concerning about it because PartitionRequests are sent and processed 
by TMs in a distributed way, unlike TDD shipping which relies on the master 
node.

To further eliminating concerns, I did an experiment to see the job execution 
time with/without the proposed change. The testing job is a word count job of 
no data, parallelism=1000, run 3 times for each case. The average execution 
time with the change is 7.4s (7.6,7.5,7.0). The average execution time without 
the change is 7.9s(8.7,6.9,8.0). Therefore, I think the change is acceptable.

> Refactor the ExecutionAttemptID to consist of ExecutionVertexID and 
> attemptNumber
> ---------------------------------------------------------------------------------
>
>                 Key: FLINK-17295
>                 URL: https://issues.apache.org/jira/browse/FLINK-17295
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Runtime / Coordination
>            Reporter: Yangze Guo
>            Assignee: Zhu Zhu
>            Priority: Major
>              Labels: pull-request-available
>
> Make the ExecutionAttemptID being composed of (ExecutionVertexID, 
> attemptNumber).



--
This message was sent by Atlassian Jira
(v8.20.7#820007)

Reply via email to