You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Zhu Zhu (Jira)" <ji...@apache.org> on 2020/01/14 07:17:00 UTC
[jira] [Commented] (FLINK-15249) Improve PipelinedRegions
calculation with Union Set
[ https://issues.apache.org/jira/browse/FLINK-15249?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17014900#comment-17014900 ]
Zhu Zhu commented on FLINK-15249:
---------------------------------
Hi [~nppoly], I tried your patch with a testing topology. The result turned out to be that region building is even slower with the patch.
Here's the result (the value is region building time eplapsed in ms):
Pure master(2020-01-14):
build regions: 700
build regions: 675
build regions: 477
build regions: 447
build regions: 461
build regions: 430
build regions: 459
build regions: 482
build regions: 733
build regions: 455
Apply #10572 on master(2020-01-14):
build regions: 1298
build regions: 1384
build regions: 1029
build regions: 867
build regions: 995
build regions: 1042
build regions: 965
build regions: 1161
build regions: 1050
build regions: 1021
The testing topology is attached.
> Improve PipelinedRegions calculation with Union Set
> ---------------------------------------------------
>
> Key: FLINK-15249
> URL: https://issues.apache.org/jira/browse/FLINK-15249
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Coordination
> Reporter: Chongchen Chen
> Priority: Major
> Labels: pull-request-available
> Attachments: PipelinedRegionComputeUtil.diff
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Union Set's Merge Set cost is O(1). current implementation is O(N). the attachment is patch.
> [Disjoint Set Data Structure|[https://en.wikipedia.org/wiki/Disjoint-set_data_structure]]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)