hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sandy Ryza (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-1899) Capacity Scheduler Usage History
Date Tue, 06 May 2014 20:04:26 GMT

    [ https://issues.apache.org/jira/browse/YARN-1899?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13991062#comment-13991062

Sandy Ryza commented on YARN-1899:

By "megabyte-seconds" and "vcore-seconds", do you mean the counters that show up for MapReduce
jobs?  We have this for MapReduce, but don't yet have a generic solution for all YARN apps.

> Capacity Scheduler Usage History
> --------------------------------
>                 Key: YARN-1899
>                 URL: https://issues.apache.org/jira/browse/YARN-1899
>             Project: Hadoop YARN
>          Issue Type: Wish
>          Components: capacityscheduler
>    Affects Versions: 2.1.0-beta
>            Reporter: Yih-Yoon Lee
> Currently Resource Manager UI provides a nice overview of Capacity Scheduler configuration
in http://localhost:8088/cluster/scheduler. If we can capture a history of the data provided
in "application queues" section, it would be helpful for administrator to review each queue
usage history and tune accordingly. Since the data is already being captured to display "real
time", perhaps the data can be kept in Ganglia.

This message was sent by Atlassian JIRA

View raw message