You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-user@hadoop.apache.org by wangxiaowei <wa...@software.ict.ac.cn> on 2008/05/15 11:52:31 UTC

can I write the Outputs of map directly to HDFS?

Hi,all
  My program just wants to collect the results of the map outputs,to save the time,can I just use Mapper Class without Reducer Class? Is it like this?

  job.setMapperClass(MyMapper.class);
  //other configurations,but without job.setReducerClass();
  JobClient.runJob(job);

How can I implement it?Thank you .

Re: can I write the Outputs of map directly to HDFS?

Posted by Yi-Kai Tsai <yi...@yahoo-inc.com>.

hi

if you just need collect without sorting,
you can do setNumReduceTasks(0);

http://hadoop.apache.org/core/docs/r0.16.4/api/org/apache/hadoop/mapred/JobConf.html#setNumReduceTasks(int)

> Hi,all
>   My program just wants to collect the results of the map outputs,to save the time,can I just use Mapper Class without Reducer Class? Is it like this?
>
>   job.setMapperClass(MyMapper.class);
>   //other configurations,but without job.setReducerClass();
>   JobClient.runJob(job);
>
> How can I implement it?Thank you .
>   


-- 
Yi-Kai Tsai (cuma) <yi...@yahoo-inc.com>, Asia Regional Search Engineering.