Hello, I've just downloaded Hadoop and basically have a few questions about getting started.
I'm currently working on a study to asses Mapreduce's performance in different situations, including smaller clusters and CMP machines. So my main questions is if there's a place I can go to to find out how exactly hadoop should be set up. Also, I am aware that we can specify the number of hosts that will be available for the mapreduce operations, but in its current form, can hadoop handle assigning tasks to multiple processors within a single machine? If so, is there any documentation on how it's done? Any help on this would really be appreciated since I'm kind of lost right now. Thank a lot in advance. Ram
