James D. Myers, Ph.D.

Director, Computational Center for Nanotechnology Innovations, Rensselaer Polytechnic Institute

Delivering Competitiveness with High Performance Computing

It is hard to argue against the value of high performance computing (HPC) as a key tool for academic discovery and industrial innovation – except by pointing out how limited its use is relative to its potential. In industry, this seeming paradox – a high-value technique being underutilized – is being described as the ‘missing middle' and the argument is made that access to HPC must be increased to support industrial competitiveness. Unfortunately, ‘increasing access' quickly gets caricatured as a need to increase the visibility of past HPC successes and to increase the provisioning of computing resources. In reality, ‘access' is much more complex, and a much broader change in the in the operational model and technical services provided by computing centers will be needed to expand academic and industrial use of HPC. In this talk, I will discuss the types of challenges that lower the value of HPC for academic and industry users and that therefore slow its adoption. I will also discuss the role of e-Science-related technologies and lessons-learned from scientific Cyberinfrastructure projects in meeting these challenges and argue that delivering competitiveness for research and design will require a combination of these technologies and innovative interaction/business models with traditional HPC capabilities.


Dr. Myers is the Director of the Computational Center for Nanotechnology Innovations (CCNI) and Professor of Practice in Computer Science at Rensselaer Polytechnic Institute. He received his B.A. in Physics from Cornell University (1985) and his Ph.D. in Chemistry from the University of California at Berkeley (1993). He has nearly two decades of experience in the development and deployment of advanced Cyberinfrastructure for research, education, and industrial application. Myers is also a co-PI on a recent NSF Major Research Instrumentation award that will more than double CCNI's computing resources as well as an NSF DataNet project exploring the integration of data-intensive computing with long-term data curation and preservation.

Workshop Program
updated: 2011-10-19