CACR Seminar: “General purpose GPU programming by CUDA—an introductory tutorial”

Tuesday November 27, 2012
Powell Booth Room 100
1:30PM

“General purpose GPU programming by CUDA—an introductory tutorial”
Dr. Hailiang Zhang, Caltech Center for Advanced Computing Research (CACR)

Abstract:
In recent years, various fields of large-scale scientific computations
have greatly benefit from the massively parallel programming.  This talk
presents a brief introduction and tutorial on the state-of-the-art
general-purpose GPU programming platform—CUDA.  The GPU device
architecture, memory hierarchy, and the general CUDA programming model
will be introduced.  The CUDA numerical schemes of some vector and matrix
operations will be demonstrated as examples.  Some CUDA applications on
molecular and biophysical modeling will be presented.  The standard CUDA
toolkit libraries and some third party APIs will also be introduced.