Long Road to Reconfigurable Supercomputing

2009-09-01 fpga pub

Keith Underwood got asked to convert one of his talks about our FPGA work into a journal article for ACM Transactions on Reconfigurable Technology and Systems. One thing that's in this paper that's particularly useful is a report card on areas where FPGA's need help in order to be successful for HPC. I think there are a lot more issues, but it's a good start for the community.

Abstract

The field of high performance computing (HPC) currently abounds with excitement about the potential of a broad class of things called accelerators. And, yet, few accelerator based systems are being deployed in general purpose HPC environments. Why is that? This article explores the challenges that accelerators face in the HPC world, with a specific focus on FPGA based systems. We begin with an overview of the characteristics and challenges of typical HPC systems and applications and discuss why FPGAs have the potential to have a significant impact. The bulk of the article is focused on twelve specific areas where FPGA researchers can make contributions to hasten the adoption of FPGAs in HPC environments.

Publications

TORTS Paper Keith D. Underwood, K. Scott Hemmert, and Craig D. Ulmer, "From Silicon to Science: The Long Road to Production Reconfigurable Supercomputing", ACM Transactions on Reconfigurable Technology and Systems, Vol. 2, No. 4, Article 26, September 2009.