Mailing-List: contact mapreduce-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: mapreduce-issues@hadoop.apache.org
Message-ID: <22708499.493611279704171051.JavaMail.jira@thor>
Date: Wed, 21 Jul 2010 05:22:51 -0400 (EDT)
From: "Hong Tang (JIRA)" <jira@apache.org>
To: mapreduce-issues@hadoop.apache.org
Subject: [jira] Updated: (MAPREDUCE-1936) [gridmix3] Make Gridmix3 more
 customizable.
In-Reply-To: <2986197.338151278979911310.JavaMail.jira@thor>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/MAPREDUCE-1936?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Hong Tang updated MAPREDUCE-1936:
---------------------------------

    Status: Patch Available  (was: Open)

New patch that addresses Chris's comments. The usage output now looks like the following:

{noformat}
Usage: gridmix [-generate <MiB>] [-users URI] [-Dname=value ...] <iopath> <trace>
  e.g. gridmix -generate 100m foo -
Configuration parameters:
   General parameters:
       gridmix.output.directory                         : Output directory
       gridmix.client.submit.threads                    : Submitting threads
       gridmix.client.pending.queue.depth               : Queued job desc
       gridmix.user.resolve.class                       : User resolution class
       gridmix.job.type                                 : Job types (LOADJOB|SLEEPJOB)
   Parameters related to job submission:
       gridmix.job-submission.default-queue             : Default queue
       gridmix.job-submission.use-queue-in-trace        : Enable/disable using queues in trace
       gridmix.job-submission.policy                    : Job submission policy (REPLAY|STRESS|SERIAL)
   Parameters specific for LOADJOB:
       gridmix.key.fraction                             : Key fraction of rec
   Parameters specific for SLEEPJOB:
       gridmix.sleep.maptask-only                       : Whether to ignore reduce tasks
       gridmix.sleep.fake-locations                     : Number of fake locations for map tasks
       gridmix.sleep.max-map-time                       : Maximum map task runtime in mili-sec
       gridmix.sleep.max-reduce-time                    : Maximum reduce task runtime in mili-sec (merge+reduce)
   Parameters specific for STRESS submission throttling policy:
       gridmix.throttle.jobs-to-tracker-ratio           : jobs vs task-tracker ratio
       gridmix.throttle.maps.task-to-slot-ratio         : maps vs map-slot ratio
       gridmix.throttle.reduces.task-to-slot-ratio      : reduces vs reduce-slot ratio
       gridmix.throttle.maps.max-slot-share-per-job     : map-slot share per job
       gridmix.throttle.reducess.max-slot-share-per-job : reduce-slot share per job
{noformat}

> [gridmix3] Make Gridmix3 more customizable.
> -------------------------------------------
>
>                 Key: MAPREDUCE-1936
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-1936
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>          Components: contrib/gridmix
>            Reporter: Hong Tang
>            Assignee: Hong Tang
>         Attachments: mr-1936-20100715.patch, mr-1936-20100720.patch, mr-1936-yhadoop-20.1xx.patch
>
>
> I'd like to make gridmix3 more customizable. Specifically, the proposed customizations include:
> - add (random) location information for each sleep map task.
> - make the parameters used in stress submission load throttling configurable.

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