The OSCAR cluster constructed at The Department of Computer Science (SIUC), is a high performance Linux cluster that is available for students and faculty for their parallel/distributed computing needs.

The present cluster is a 9-node Red Hat based cluster implimented using OSCAR (Open Source Cluster Application Resource) software package. The head node is a powerful Pentium IV Linux box while the other 8 nodes are identical 2 GHz Linux boxes. OSCAR was chosen over Beowulf due to the ease of installation as well as its comprehensive package which includes many compilers and application softwares. One of the main package that is being used on OSCAR is LAM-MPI, a popular implementation of the MPI parallel programming paradigm. LAM-MPI can be used for developing parallel programs and it is one of the leading implementations available. Another useful program is the Portable Batch System (PBS) which can be used for scheduling parallel jobs.

Running parallel programs on the cluster is a straight-forward job that can be understood by looking at the user guide provided. If your application demands very intensive use of cluster resources, you should contact the cluster administrator before proceeding with your program. More information regarding the features and about using the cluster can be found in the official user guide. There is a quick start guide that is available here.

For more information, requesting account, or reporting trouble please email Raheel Ahmad or Dr. Shahram Rahimi.