Deploying a High Performance Computing Cluster at a Predominantly Undergraduate Institution
Evan C. Lemley, Tom Chen · 2018
This paper is focused on the deployment and use of a high performance computing HPC cluster at the University of Central Oklahoma, which is a predominantly undergraduate institution (PUI) and a Carnegie Master's Level -Larger Programs Classification.In January 2015, UCO received a National Science Foundation (NSF) Major Research Instrumentation (MRI) grant to fund an HPC cluster to enhance and extend computationally oriented research and education.By July of 2015 the Buddy Cluster (so named for the school mascot) was deployed and ready for early users.In September 2015 there were only a handful of users compared to January 2017 when a total of 150 users have been registered on the system.The Buddy Cluster has been used for a wide range of engineering and other applications such as bioinformatics, computational fluid dynamics, heat transfer, biomedical applications, image processing, and statistical analysis.Note the bulk of the use of Buddy is by the Department of Engineering and Physics Department at UCO.Although the topics discussed are general, the paper addresses how engineering physics programs can deploy and make use of similar clusters.This paper will discuss the impacts to computational research and the research infrastructure at UCO.The process to acquire and deploy the Buddy Cluster will be detailed including the process to solicit proposals from and choose a cluster vendor.The process to get users working in the cluster environment -including internet browser-based access and use of certain cluster software and the use of more standard command line access and use -are discussed.