hadoop-mapreduce-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From jingguo yao <yaojing...@gmail.com>
Subject Re: How to set the number of tasks running on each slave?
Date Tue, 16 Nov 2010 05:56:48 GMT
mapred.tasktracker.map.tasks.maximum: how many map tasks can be run
concurrently on a map-reduce slave node.

mapred.tasktracker.reduce.tasks.maximum: how many reduce tasks can be run
concurrently on a map-reduce slave node.

mapred.reduce.tasks: how many reduce task for a specific job.

mapred.map.tasks:  how many map task for a specific job. Note that this
configuration is just a lint.


On Tue, Nov 16, 2010 at 11:20 AM, Wei Xue <simonxue21@gmail.com> wrote:

> I depends on how many tasks you created, how many slave nodes in the
> cluster, how many slots allocated in each slave node, and how each node is
> occupied. Each of them has corresponding setting parameters. Please refer to
> hadoop document for more details.
>
> 2010/11/16 Shen LI <geminialex007@gmail.com>
>
> How to set the number of tasks running on each slave? (e.g., 4 tasks on
>> each slave node)
>>
>> Thanks
>>
>
>


-- 
Jingguo

Mime
View raw message