Community Clusters or the Cloud
Preston Smith, Stephen Lien Harrell, Alex Younts, Xiao Zhu · Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) · 2019
Understanding total cost of ownership (TCO) for HPC systems on a site-specific range of applications is a critical component of any RFP or capacity plan at research computing centers. With the increasing industry use of the HPC in the cloud, understanding and comparing costs of on-premises HPC resources and cloud-based options is more important than ever. In this paper we present methods to determine site-specific applications mix, calculating TCO for on-premises and cloud HPC resources, and the comparison of the two. Additionally we present a cost-analysis of Purdue's Community Cluster program with current cloud offerings.