You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Mason Chen (Jira)" <ji...@apache.org> on 2022/01/13 22:02:00 UTC

[jira] [Comment Edited] (FLINK-25646) Document buffer debloating issues with high parallelism

    [ https://issues.apache.org/jira/browse/FLINK-25646?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17475778#comment-17475778 ] 

Mason Chen edited comment on FLINK-25646 at 1/13/22, 10:01 PM:
---------------------------------------------------------------

Hi [~akalashnikov], thanks for the documentation! Could you attach a reference to the benchmark? I'd like to take a look.

Is the observation with unaligned and/or aligned checkpoints?


was (Author: mason6345):
Hi [~akalashnikov], thanks for the documentation! Could you attach a reference to the benchmark? I'd like to take a look

> Document buffer debloating issues with high parallelism
> -------------------------------------------------------
>
>                 Key: FLINK-25646
>                 URL: https://issues.apache.org/jira/browse/FLINK-25646
>             Project: Flink
>          Issue Type: Improvement
>          Components: Documentation, Runtime / Network
>    Affects Versions: 1.14.0
>            Reporter: Anton Kalashnikov
>            Assignee: Anton Kalashnikov
>            Priority: Major
>              Labels: pull-request-available
>
> According to last benchmarks, there are some problems with buffer debloat when job has high parallelism. The high parallelism means the different value from job to job but in general it is more than 200. So it makes sense to document that problem and propose the solution - increasing the number of buffers.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)