Par Lab Seminar: GPU Computing Architecture
John Nickolls and Michael Shebanow of NVIDIA will speak on Thursday, October 22 at 11am in 430 Soda Hall (the Woz).
GPU computing is becoming widely used in consumer PC applications and high performance parallel computing. The talk will cover new developments in NVIDIA GPU computing architecture and the CUDA parallel programming model that enable broader use of GPU computing. The new Fermi architecture provides higher performance, IEEE 754-2008 floating point for singles and doubles, faster double precision, 64-bit addressing, unified memory space addressing, a cached memory hierarchy, ECC memory, and support for C++.
John Nickolls is director of architecture at NVIDIA for GPU computing. He was previously with Broadcom, Silicon Spice, Sun Microsystems, and was a cofounder of MasPar Computer. His interests include parallel processing systems, languages, and architectures. He has a B.S. in electrical engineering and computer science from the University of Illinois, and M.S. and Ph.D. degrees in electrical engineering from Stanford University.
Michael Shebanow is a Senior Architecture Manager in the GPU Products group at NVIDIA. Prior to NVIDIA, Michael has held a number of positions in processor design at AMD, HAL Computer Systems, Cyrix, and Motorola. Michael’s technical areas of interest are in computer architecture and microarchitecture. Michael received a Ph.D. in Computer Science from the University of California, Berkeley and a B.S. in Electrical Engineering, also from Berkeley.