hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Wangda Tan (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-2498) Respect labels in preemption policy of capacity scheduler for inter-queue preemption
Date Thu, 23 Apr 2015 23:15:39 GMT

    [ https://issues.apache.org/jira/browse/YARN-2498?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14510043#comment-14510043
] 

Wangda Tan commented on YARN-2498:
----------------------------------

Thanks for comments, [~jianhe]. Addressed all comments except:
bq. below code at line 744 is dup with the check at line 650 ?
No, when there's no more resource need to preempt at the middle of looking at all non-exclusive
containers in the queue, it need return instead of continue

Attached patch ver.3

> Respect labels in preemption policy of capacity scheduler for inter-queue preemption
> ------------------------------------------------------------------------------------
>
>                 Key: YARN-2498
>                 URL: https://issues.apache.org/jira/browse/YARN-2498
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: resourcemanager
>            Reporter: Wangda Tan
>            Assignee: Wangda Tan
>         Attachments: STALED-YARN-2498.zip, YARN-2498.1.patch, YARN-2498.2.patch, YARN-2498.3.patch
>
>
> There're 3 stages in ProportionalCapacityPreemptionPolicy,
> # Recursively calculate {{ideal_assigned}} for queue. This is depends on available resource,
resource used/pending in each queue and guaranteed capacity of each queue.
> # Mark to-be preempted containers: For each over-satisfied queue, it will mark some containers
will be preempted.
> # Notify scheduler about to-be preempted container.
> We need respect labels in the cluster for both #1 and #2:
> For #1, when calculating ideal_assigned for each queue, we need get by-partition-ideal-assigned
according to queue's guaranteed/maximum/used/pending resource on specific partition.
> For #2, when we make decision about whether we need preempt a container, we need make
sure, resource this container is *possibly* usable by a queue which is under-satisfied and
has pending resource.
> In addition, we need to handle ignore_partition_exclusivity case, when we need to preempt
containers from a queue's partition, we will first preempt ignore_partition_exclusivity allocated
containers.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message