AGENDA

Cloud Computing for parallel Scientific HPC Applications: Feasibility of Running Coupled Atmosphere-Ocean Climate Models on Amazon's EC2

Constantinos Evangelinos, Chris Hill

Abstract: In this article we describe the application of HPC standard benchmark tests to Amazon's EC2 cloud computing system, in order to explore the utility of EC2 for modest HPC style applications. Based on these benchmarks we find that the EC2 cloud system is emerging as a credible solution for supporting responsive on-demand, small sized, HPC applications. We illustrate this with a demonstration of a low-order coupled atmosphere-ocean simulation running in parallel on an EC2 system. This demonstration highlights one significant way in which cloud computing could impact traditional HPC computing paradigms. Taken together with related ease-of-use technologies (described only briefly in this paper), systems like EC2 could start to shift the manner in which many HPC systems are deployed.