ABSTRACT

Computing is in crisis. Uniprocessor throughput has increased only slowly over the last ten years; yet developing efficient algorithms for multi-core processors remains ad hoc and frequently fails. Latency is no longer improving either - because multiple cores now compete for limited, off-chip memory bandwidth, and each core needs access to lots of off - chip memory. GPUs offer higher performance on some tasks, but high-performance GPU code is painfully hardware-specific, and writing it is a black art. FPGAs are still harder to program and to optimize but hold the theoretical promise of even higher performance and energy efficiency. Can we do better? Can we build a better computer that combines the large -scale spatial performance of FPGAs and GPUs with the temporal simplicity of the CPU? Can we make that computer much easier to program than current, massively parallel devices? This talk will describe one possible path forward.

BIOGRAPHY

Steve Teig is the President and Chief Technology Officer of Tabula and the inventor of Tabula"s Spacetime 3-Dimensional Programmable Logic Architecture. In 2012, Tabula has been recognized with an Edison Award, an American Business Gold Award, inclusion in MIT"s TR50 list of the 50 most innovative companies worldwide, and as #3 in the Wall Street Journal"s "Next Big Thing" list of the most promising venture-backed startup companies. Prior to founding Tabula, Steve was CTO of Cadence Design Systems, having joined Cadence through its acquisition of Simplex Solutions, where he was also CTO. Prior to Simplex, he was CTO of two successful biotechnology companies: CombiChem and BioCAD. Earlier in his career, he developed foundational logic simulation and place-and-route technologies that continue to have far-reaching influence. Steve received a B.S.E. degree in Electrical Engineering and Computer Science from Princeton University and holds 282 patents. In 2011, he was awarded the World Technology Award for IT hardware innovation.

Conference information