You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Shammon (Jira)" <ji...@apache.org> on 2022/01/14 02:28:00 UTC

[jira] [Updated] (FLINK-25329) Improvement of execution graph store in flink session cluster for jobs

     [ https://issues.apache.org/jira/browse/FLINK-25329?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Shammon updated FLINK-25329:
----------------------------
    Summary: Improvement of execution graph store in flink session cluster for jobs  (was: Improvement of disk operations in flink session cluster for jobs)

> Improvement of execution graph store in flink session cluster for jobs
> ----------------------------------------------------------------------
>
>                 Key: FLINK-25329
>                 URL: https://issues.apache.org/jira/browse/FLINK-25329
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Runtime / Coordination
>    Affects Versions: 1.14.0, 1.12.5, 1.13.3
>            Reporter: Shammon
>            Priority: Major
>
> Flink session cluster uses files to store info of jobs after they reach termination with `FileExecutionGraphInfoStore`, each job will generate one file. When the cluster executes many small jobs concurrently, there will be many disk related operations, which will
> 1> Increase the CPU usage of `Dispatcher`
> 2> Decrease the performance of the jobs in the cluster.
> We hope to improve the disk operations in `FileExecutionGraphInfoStore` to increase the performance of session cluster.



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