Hi !
Hm, a few months later...
I managed to build the cluster and get Gaussian 03 to run on it. However, I used a cluster installation toolkit program suite named OSCAR which makes life much easier for chemists who aren't highly knowledgeable in Linux.
In the process of setting up the cluster, I found the following resources very helpful:
Cluster Resources Incorporated, 2009 Torque Resource Manager [Online] Available from
http://www.clusterresources.com/products/torque/docs/a.ltorquequickstart.shtmlHekman J.P., 1997 Linux in a Nutshell Sebastopol, CA, O'Reilly & Associates
Sloan J.D., 2005 High Performance Linux Clusters with OSCAR, Rocks, openMosix & MPI Sebastopol, CA, O'Reilly Media.
Van Eck A., 2009a Welcome to the Computational Chemistry SIG – Software Web [Online] South Africa, University of the Free State, Available from
http://sig.sagrid.ac.za/bin/view/CompChem/Software/OscarInstallIf you want to do something similar I would suggest that you do thorough (and I mean
thorough) reading before you start loading any operating systems or application software on your hardware. Even the hardware needs special attention actually.
I want to encourage anyone who needs computational chemistry results in less time without sacrificing accuracy, and who doesn't have a big budget, to consider this option. I think any chemist would be able to do this, however, the time it would take to build a cluster is directly proportional to your computer background knowledge.
;-)