Programming finite-difference time-domain for graphics processor units using compute unified device architecture

Author

Demir, Veysel ; Elsherbeni, Atef Z.

Author_Institution

Dept. of Electr. Eng., Northern Illinois Univ., DeKalb, IL, USA

fYear

2010

fDate

11-17 July 2010

Firstpage

1

Lastpage

4

Abstract

Recently graphic processing units (GPU´s) have become the hardware platforms to perform high performance scientific computing them. The unavailability of high level languages to program graphics cards had prevented the widespread use of GPUs. Relatively recently Compute Unified Device Architecture (CUDA) development environment has been introduced by NVIDIA and made GPU programming much easier. This contribution presents an implementation of finite-difference time-domain (FDTD) method using CUDA. A thread-to-cell mapping algorithm is presented and performance of this algorithm is provided.

Keywords

computer graphic equipment; coprocessors; finite difference time-domain analysis; programming environments; CUDA development environment; NVIDIA; compute unified device architecture; finite-difference time-domain programming; graphic processor units; high-level languages; program graphics cards; thread-to-cell mapping algorithm; Algorithm design and analysis; Computer architecture; Equations; Finite difference methods; Graphics; Graphics processing unit; Time domain analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Antennas and Propagation Society International Symposium (APSURSI), 2010 IEEE

Conference_Location

Toronto, ON

ISSN

1522-3965

Print_ISBN

978-1-4244-4967-5

Type

conf

DOI

10.1109/APS.2010.5562117

Filename

5562117