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

Posted to mapreduce-dev@hadoop.apache.org by Sonal Kumar <so...@oracle.com> on 2011/08/25 15:38:31 UTC

Regarding number of map tasks

Hi,

Just want to ask , how one can control the number of mappers for a given job.
I have a 2 cpu with 2 cores machine i.e 4 cores having Red Hat 4.1.2-48.
I have a sqoop export job, with
Input size :320,139,321Bytes
Block size : 4096Bytes

It is creating 3 map tasks for this export job.

When I am passing --num-mappers 4, still it is creating 3 splits and assigning 3 mappers to it.

Console out put can be viewed from :http://pastebin.com/KVPp5Zjp

I went through the code for CombineFileInputFormat.java , but I am not able to figure out , the reason for this behaviour.

Can you help in this?


Thanks & Regards,
Sonal Kumar