hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sunil G (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-4161) Capacity Scheduler : Assign single or multiple containers per heart beat driven by configuration
Date Wed, 28 Jun 2017 21:34:00 GMT

    [ https://issues.apache.org/jira/browse/YARN-4161?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16067270#comment-16067270

Sunil G commented on YARN-4161:

Thanks [~ywskycn] for rebasing the patch.
Few high level comments
# Could you please name patch to something like YARN-4161.002.patch etc by following the naming
convention. It will be more easier.
# I thoughts we can rename the configuration to "yarn.scheduler.capacity.per-node-heartbeat.maximum-container-assignments".
More thoughts are welcome
# Once 2 is confirmed, rename all variables and getters/setter in align with config name
# I prefer not to have  a config like "yarn.scheduler.capacity.assignmultiple". If we can
make the default of "yarn.scheduler.capacity.per-node-heartbeat.maximum-container-assignments"
to -1, we can assume the feature is off and we can continue allocate any numbers of containers
as possible in a heartbeat. Once configured, we can assume that feature is on and restrict
as per the limit imposed.

> Capacity Scheduler : Assign single or multiple containers per heart beat driven by configuration
> ------------------------------------------------------------------------------------------------
>                 Key: YARN-4161
>                 URL: https://issues.apache.org/jira/browse/YARN-4161
>             Project: Hadoop YARN
>          Issue Type: New Feature
>          Components: capacity scheduler
>            Reporter: Mayank Bansal
>            Assignee: Mayank Bansal
>              Labels: oct16-medium
>         Attachments: YARN-4161.patch, YARN-4161.patch.1
> Capacity Scheduler right now schedules multiple containers per heart beat if there are
more resources available in the node.
> This approach works fine however in some cases its not distribute the load across the
cluster hence throughput of the cluster suffers. I am adding feature to drive that using configuration
by that we can control the number of containers assigned per heart beat.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org

View raw message