You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Flink Jira Bot (Jira)" <ji...@apache.org> on 2021/11/09 22:39:00 UTC

[jira] [Updated] (FLINK-15848) Support both fixed allocator and dynamic allocator in flink

     [ https://issues.apache.org/jira/browse/FLINK-15848?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Flink Jira Bot updated FLINK-15848:
-----------------------------------
    Labels: auto-deprioritized-major stale-minor  (was: auto-deprioritized-major)

I am the [Flink Jira Bot|https://github.com/apache/flink-jira-bot/] and I help the community manage its development. I see this issues has been marked as Minor but is unassigned and neither itself nor its Sub-Tasks have been updated for 180 days. I have gone ahead and marked it "stale-minor". If this ticket is still Minor, please either assign yourself or give an update. Afterwards, please remove the label or in 7 days the issue will be deprioritized.


> Support both fixed allocator and dynamic allocator in flink
> -----------------------------------------------------------
>
>                 Key: FLINK-15848
>                 URL: https://issues.apache.org/jira/browse/FLINK-15848
>             Project: Flink
>          Issue Type: Improvement
>          Components: Runtime / Task
>    Affects Versions: 1.10.0
>            Reporter: liupengcheng
>            Priority: Minor
>              Labels: auto-deprioritized-major, stale-minor
>
> Currently, we removed static allocator and only support dynamic allocation in flink1.10, however, this allocator still has some drawbacks:
>  # Can not allocate resources in a range, which means the resource usage is not under control, this has very bad effect in a shared resource cluster(e.g. Yarn), one big query or job may occupy all the resources and  cause other jobs block.
>  # Not support static resource allocation. That means we can hardly do benchmark testing across engines(e.g. Spark). Also, in resource shared cluster(e.g. Yarn) that support over-allocation, it's hard to align the resources usage.
> As discussed in FLINK-12362 ,  we should support both fixed allocator and dynamic allocator(dynamically allocating in a range) in flink.
>  



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