Systems Seminar: Prof. William J. Dally, EE and CS Departments, Stanford University

A Bandwidth-Efficient Architecture for Media Processing

Abstract

Media applications are characterized by large amounts of available parallelism, little data reuse, and a high computation to memory access ratio. While these characteristics are poorly matched to conventional architectures, they are a good fit for modern VLSI technology with its high arithmetic capacity but limited global bandwidth. The stream programming model, in which an application is coded as streams of data records passing through computation kernels, facilitates mapping of applications onto VLSI architectures by exposing both parallelism and locality in the application. The Imagine architecture supports the stream programming model by providing a bandwidth hierarchy tailored to the demands of media applications. Compared to a conventional scalar processor, Imagine reduces the global register and memory bandwidth required by typical applications by factors of 21 and 62 respectively. This bandwidth efficiency enables a single chip Imagine processor to achieve a peak performance of 20GFLOPS (32-bit FP) and sustained performance of up to 7.5GOPS on media processing kernels.

Speaker Bio

William Dally received the B.S. degree in Electrical Engineering from Virginia Polytechnic Institute, the M.S. degree in Electrical Engineering from Stanford University, and the Ph.D. degree in Computer Science from Caltech. Bill and his group have developed network architecture, signaling, routing, and synchronization technology that can be found in most large parallel computers today. While at Bell Telephone Laboratories Bill contributed to the design of the BELLMAC32 microprocessor and designed the MARS hardware accelerator. He was a Research Assistant and then a Research Fellow at Caltech where he designed the MOSSIM Simulation Engine and the Torus Routing Chip which pioneered wormhole routing and virtual-channel flow control. While a Professor of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology he and his group built the J-Machine and the M-Machine, experimental parallel computer systems that pioneered the separation of mechanisms from programming models and demonstrated very low overhead mechanisms for synchronization and communication. Bill has worked with several manufacturers of parallel computers including Cray Research and Intel to incorporate many of these innovations in commercial machines. Bill is currently a Professor of Electrical Engineering and Computer Science at Stanford University where he leads projects on high-speed signaling, multiprocessor architecture, and graphics architecture. He has published over 80 papers in these areas and is an author of the textbook, Digital Systems Engineering.