Well to start my name is William and I am an undergraduate at UW-Oshkosh studying Microbiology. The inception of the cluster was due to computational needs presented in the MAANOVA (MicroArray ANalysis Of VAriace) package under the R Statistical Package. We will be primarily using the cluster to do analysis on arabadopsis under Dr. Dorn.
When I started the project is did searches using Google to try to find setup procedures, tips and tricks, and any other useful information on building our setup. There are many websites out there on how to do it, though dated, they were useful as I was able to peice together the idea behind how to build a beowulf cluster, but I then used the Debian HOWTO pages to fill in the gaps and build my own practical guide of installation. A warning to those of you out there who have not worked with linux before: Don't give up!
If you are new to Linux that is ok, and this is going be an excercise in trial and error more than anything else. I personally reinstalled my operating system about 5 times before I got it the way it needed to be to work.
To start, some helpful websites that I visited...
As I said before, none of the above give a start to finish cluster setup, many of them expect you to know / have worked extensively in linux before; however this blog is going to be a setup and administration guide for someone new to linux but with some computer background. I will do the best I can to point you in the right direction, but the only way to succeed is trail and error and good old fashioned persistance.
I recommend that you look at the above websites, a cursory glance at least, so you will get a better understanding as to my design choices and how I put together the cluster here.
The next post will be the specs (both sofware and hardware) that we are using for our cluster and the minimum recommendation for what you will need to get started.