Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Proceedings of The National Conference On Undergraduate Research (NCUR) 2006

Summary: Proceedings of The National Conference
On Undergraduate Research (NCUR) 2006
The University of North Carolina at Asheville
Asheville, North Carolina
April 6 8, 2006
Using a Graphics Processor Unit (GPU) for Feature Extraction from
Turbulent Flow datasets
Andrew Davidson
Computer Engineering and
Center for Computation and Technology
Louisiana State University
Johnston Hall
Baton Rouge, Louisiana 70809 USA
Faculty Advisor: Gabrielle Allen
Due to its high cost effectiveness (Mega-flops/dollar), the GPU has become a tempting co-processing unit.
However, the programmability on a GPU has remained a challenge. There is little flexibility in the current
languages, there are many hardware constraints, and CPU code cannot be converted directly into GPU
code. Much progress is being made towards effective GPU co-processing. New graphic languages such as
Cg, Brook, GLSL, and HLSL have allowed programmers more control and options on the GPU. Using


Source: Allen, Gabrielle - Department of Computer Science, Louisiana State University


Collections: Computer Technologies and Information Sciences