Hello.

Am 14.09.2012 um 08:03 schrieb Jason Yang <[email protected]>:

> I have a question about how does the pseudo-distributed Hadoop cluster work:
> 
> As many map tasks are submitted to the pseudo-distributed Hadoop cluster, 
> does the hadoop run each mapper in sequence ? or does it run these mappers in 
> different threads or something could be parallel?

pseudo-distributed mode is a one node cluster. You have a namenode, a 
jobtracker, and a single datanode and tasktracker running. You can verify with 
"jps" command.

The default setting is that a tasktracker can run up to two map and reduce 
tasks in parallel (mapred.tasktracker.map.tasks.maximum and 
mapred.tasktracker.reduce.tasks.maximum), so you will actually see some 
concurrency on your one machine.

Kai

-- 
Kai Voigt
[email protected]




Reply via email to