hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Arun C Murthy (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-5964) Fix the 'cluster drain' problem in the Capacity Scheduler wrt High RAM Jobs
Date Thu, 18 Jun 2009 07:44:07 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-5964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12721114#action_12721114
] 

Arun C Murthy commented on HADOOP-5964:
---------------------------------------

Some notes about this patch:

# I've introduced a new org.apache.hadoop.mapred.server.jobtracker.TaskTracker class to track
all information about a given TaskTracker at the JobTracker such as TaskTrackerStatus, reservations
for high-ram jobs etc. I chose the new package based on the proposals at HADOOP-398.
# I've changed the TaskScheduler.assignTasks api to use the newly introduced rather than the
tepid TaskTrackerStatus. Clearly other schedulers (ala CapacityTaskScheduler) can start to
take advantage of this as I've fixed them appropriately in this patch.
# I've had to make some classes public (JobInProgress, TaskTrackerStatus) for org.apache.hadoop.mapred.server.jobtracker.TaskTracker
to work with appropriate caveats in the javadocs etc.

> Fix the 'cluster drain' problem in the Capacity Scheduler wrt High RAM Jobs
> ---------------------------------------------------------------------------
>
>                 Key: HADOOP-5964
>                 URL: https://issues.apache.org/jira/browse/HADOOP-5964
>             Project: Hadoop Core
>          Issue Type: Bug
>          Components: contrib/capacity-sched
>    Affects Versions: 0.20.0
>            Reporter: Arun C Murthy
>            Assignee: Arun C Murthy
>             Fix For: 0.21.0
>
>         Attachments: HADOOP-5964_0_20090602.patch, HADOOP-5964_1_20090608.patch, HADOOP-5964_2_20090609.patch,
HADOOP-5964_4_20090615.patch, HADOOP-5964_6_20090617.patch, HADOOP-5964_7_20090618.patch
>
>
> When a HighRAMJob turns up at the head of the queue, the current implementation of support
for HighRAMJobs in the Capacity Scheduler has problem in that the scheduler stops assigning
tasks to all TaskTrackers in the cluster until a HighRAMJob finds a suitable TaskTrackers
for all its tasks.
> This causes a severe utilization problem since effectively no new tasks are allowed to
run until the HighRAMJob (at the head of the queue) gets slots.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message