You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Matthias (Jira)" <ji...@apache.org> on 2020/10/23 10:15:00 UTC
[jira] [Commented] (FLINK-11815) Support execution with static
resources (Flip-6): E.g. spreading tasks across available set of TMs
[ https://issues.apache.org/jira/browse/FLINK-11815?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17219617#comment-17219617 ]
Matthias commented on FLINK-11815:
----------------------------------
[~pnowojski] [~trohrmann] Does this issue still need to be kept open considering that the (only?) subtask is solved? It looks like it is idling for quite some time...
> Support execution with static resources (Flip-6): E.g. spreading tasks across available set of TMs
> --------------------------------------------------------------------------------------------------
>
> Key: FLINK-11815
> URL: https://issues.apache.org/jira/browse/FLINK-11815
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Coordination
> Affects Versions: 1.6.4, 1.7.2, 1.8.0
> Reporter: Piotr Nowojski
> Priority: Major
>
> FLIP-6 code path changed how tasks are distributed across the cluster which is causing problems in some scenarios. This issue intends to improve this.
> Known issues:
> 1. key groups are currently assigned randomly, ignoring their size. Previously larger key groups were evenly spread around the available task managers
> 2. if task slots are registered before slot request, the code have a tendency to group requests together on the same machine because we are using a LinkedHashMap
> 3. ???
> CC [~trohrmann@apache.org]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)