> Electrical and Computer Engineering > Letter from the Chair > Distinguished Lectures Series 2007-2008 > Wen-Mei Hwu

Wen-Mei Hwu

hwu, Wen-mei Hwu's photo, DLS 2007

"GPU Computing: Why is it exciting to many application developers?"

Thursday, November 29, 2007, 3 p.m
Sandford Fleming Building, Room 1105

Abstract: Modern GPUs such as the NVIDIA GeForce-8 series are increasingly designed as massively parallel programmable processors. New architecture interfaces have alleviated the need for application developers to deal with graphics programming languages and interfaces. For example, CUDA programmers simply treat the GeForce 8800 GTX as a computing processor that consists of 128 processor cores, has a peak performance of 367 single-precision GFLOPS, features 86.4 GB/s memory bandwidth, contains 768MB of main memory, incurs very little cost in creating thousands of threads, and allows efficient data sharing and synchronization across subsets of threads. With millions of units already in use, it has also become arguably the largest installation of massively parallel system in history.

In the next decade, we are going to see continued performance scaling in this type of massively parallel compute engines. According to the semiconductor industry roadmap, these accelerators could provide up to 10,000x speedup over our current microprocessors by the end of the year 2016. Such a dramatic increase in computation power will likely enable revolutionary work in science, engineering and many other disciplines. Like any other massively parallel computer system, in order to achieve high performance, an application programmer currently has to understand the desirable parallel programming idioms, potential performance pitfalls, and proven coding strategies for the platform. However, the programming and code optimization models of GPU computing design are quite different from those of traditional CPUs. In this presentation, I will describe the vision and recent results of a collaborative effort between the University of Illinois and NVIDIA on building an infrastructure of programming tools, educational materials, application development experience, and architectural directions needed for application developers to fully exploit the hardware compute power of current and future GPU computing platforms.

Bio: Wen-mei W. Hwu is a Professor and holds the Sanders-AMD Endowed Chair in the Department
of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign. His research interests are in the area of architecture, implementation, and software for high performance computer systems.

Dr. Hwu is the director of the IMPACT research group (www.crhc.uiuc.edu/Impact). For his contributions in research and teaching, he received the ACM SigArch Maurice Wilkes Award, the ACM Grace Murray Hopper Award, the Tau Beta Pi Daniel C. Drucker Eminent Faculty Award, and ISCA Most Influential Paper Award.

Dr. Hwu is a fellow of IEEE and ACM. He serves on the Executive Committee of the MARCO/DARPA C2S2 (www.c2s2.org) and GSRC (www.gigascale.org) Focus Research Centers. He leads the GSRC Concurrent Systems Theme with Kurt Keutzer. He also serves on the GELATO Strategy Council (www.gelato.org). Dr. Hwu received his Ph.D. degree in Computer Science from the University of California, Berkeley.

Wen-Mei Hwu

"GPU Computing: Why is it exciting to many application developers?" Thursday, November 29, 2007, 3 p.mSandford Fleming Building, Room 1105

"GPU Computing: Why is it exciting to many application developers?"

Thursday, November 29, 2007, 3 p.m
Sandford Fleming Building, Room 1105