Upcoming Event:
Date: October 1, 2008
6:00pm: Refreshments, Food, and Social Time
7:00pm: Technical Presentation
Place: Simularium Room at the University of
California in Santa Cruz (Room 180, Engineering Bldg 2)
Follow directions and parking Information. There will
be a two dollar fee for parking.
RSVP: email to siero@ee.com
Title
Parallel Computing with CUDA on GPUs
Speaker
Dr. John Nickolls & Dr. Massimiliano Fatica, Nvidia Corporation
Abstract
The CUDA parallel programming model provides a straightforward means of describing inherently parallel computations. It lets programmers write scalable parallel programs with a minimal extension of the C language. NVIDIA's Tesla GPU architecture delivers high computational throughput on massively parallel problems. This talk describes how to write CUDA C programs, outlines the parallel computing architecture of a 240-processor GPU, and surveys applications of CUDA to different problems and the parallel speedups obtained on GPUs over traditional sequential CPU codes.
Biography
John Nickolls is director of architecture at NVIDIA for GPU computing. He was previously with Broadcom, Silicon Spice, Sun Microsystems, and was a co-founder of MasPar Computer. His interests include parallel processing systems, languages, and architectures. Nickolls has a BS in electrical engineering and computer science from the University of Illinois, and MS and PhD degrees in electrical engineering from Stanford University.
Massimiliano Fatica is a Senior Applied Engineer at NVIDIA where he works in the area of GPU computing (high-performance computing and clusters). Prior to joining NVIDIA, he was a research staff member at Stanford University where he worked on applications for the Stanford Streaming Supercomputer. He holds a laurea in Aeronautical Engineering and a PhD in Theoretical and Applied Mechanics from the University of Rome "La Sapienza".