Enhanced Parallel FDTD Method Using SSE Instruction Sets

Authors

  • Lihong Zhang School of Information Engineering Communication University of China, Beijing, 100024, China
  • Xiaoling Yang Penn State University University Park, PA, 16802, USA
  • Wenhua Yu Penn State University University Park, PA, 16802, USA

DOI:

https://doi.org/10.13052/aces/2156-3306.2711

Keywords:

FDTD method

Abstract

To accelerate the simulation of the parallel FDTD method, this paper proposes an effective hardware acceleration technique based on the SSE instruction sets, and puts forward a three-level data parallel algorithm based on MPI, OpenMP and SSE instructions. To demonstrate the acceleration effect of this technique, this paper develops two types of codes using C language: one is based on MPI + OpenMP, another is based on MPI + OpenMP + SSE, and then draws a comparison between the computing time of the two types of codes in the numerical experiments for the same electromagnetic radiation problems. The experimental results show that this acceleration technique can achieve an acceleration rate of 2.44 for the ideal case on a PC cluster and 2.37 for the practical problem on a 2-CPU workstation without requiring any extra hardware investment, and provide an efficient and economical technique for the electromagnetic simulations.

Downloads

Download data is not yet available.

Author Biographies

Lihong Zhang, School of Information Engineering Communication University of China, Beijing, 100024, China

Lihong Zhang is presently
working on her Ph.D.in parallel
computing and will graduate next
year from Communication
University of China. Her research
interests include parallel processing
techniques, numerical methods and
software development.

Xiaoling Yang, Penn State University University Park, PA, 16802, USA

Xiaoling Yang is a research
associate in Material Research
Institute of Pennsylvania State
University. He received his B.S.
and M.S. in Communication and
Mathematics from Tianjin
University in 2001 and 2004,
respectively. He has published
three books related to the FDTDmethod, parallel
processing techniques, software development
technique, and simulation techniques. He has published
over 20 technical papers. His research interests include
numerical methods, visual languages and software
development.

Wenhua Yu, Penn State University University Park, PA, 16802, USA

Wenhua Yu is with 2COMU, Inc.
and serves as the president of
2COMU. He was with
Pennsylvania State University from
1996 to 2011. He received his
Ph.D. in Electrical Engineering
from the Southwest Jiaotong
University in 1994. He worked at
the Beijing Institute of Technology as a Postdoctoral
Research Associate from February 1995 to August
1996. He has published six books related to the FDTD
method, parallel processing techniques, software
development technique, and simulation techniques from
2003 to 2011. He has published over 150 technical
papers and four book chapters. He is a senior member
of IEEE. His research interests include computational
electromagnetic methods, software development
techniques, parallel processing techniques, and
simulation and design of the antennas, antenna arrays
and microwave circuits.

References

K. S. Yee, “Numerical Solution of Initial Boundary

Value Problems Involving Maxwell ́s Equations in

Isotropic Media,” IEEE Trans. Antennas

Propagat., vol. AP-14, pp. 302-307, 1966.

A. Taflove and S. Hagness,Computational

Electrodynamics: The Finite-Difference Time

Domain Method, Artech House, Norwood, May

W. Yu, R. Mittra, T. Su, et al.,Parallel Finite

Difference Time Domain Method, Communication

University of China Press, July, 2005.

W. Yu, et al., “A Robust Parallel Conformal FDTD

Processing Package using the MPILibrary,” IEEE

Antennas and Propagation Magazine, vol. 47, no.

, pp. 39-59, June 2005.

Y. Zhang, W. Ding, and C. H. Liang, “Study on the

Optimum Virtual Topology for MPI Based Parallel

Conformal FDTD Algorithm on PC Clusters,” J. of

Electromagn. Waves and Appl., vol. 19, no. 13, pp.

-1831, 2005.

W. Gropp, E. Lusk, and A. Skjellum,Using MPI:

Portable Parallel Programming with the

Message-Passing Interface. 2nd ed., MIT Press,

Cambridge, Nov., 1999.

https://computing.llnl.gov/tutorials/openMP/

V. Demir and A. Z. Elsherbeni, “Compute Unified

Device Architecture (CUDA) based Finite-

Difference Time-Domain (FDTD)

Implementation,” Applied Computational

Electromagnetic Society (ACES) Journal, vol. 25,

no. 4, pp. 303-314, Apr. 2010.

N. Takada, T. Shimobaba, N. Masuda, and T. Ito,

“Improved Performance of FDTDComputation

using a Thread Block Constructed as aTwo-

Dimensional Array with CUDA,”Applied

Computational Electromagnetic Society (ACES)

Journal, vol. 25, no. 12, pp. 1061-1069, Dec.

M. Ujaldon, “Using GPUs forAccelerating

Electromagnetic Simulations,” Applied

Computational Electromagnetic Society (ACES)

Journal, vol. 25, no. 4, pp. 294-302, Apr. 2010.

M. Weldon, L. Maxwell, D. Cyca, M. Hughes, C.

Whelan, and M. Okoniewski, “APractical Look

at GPU-Accelerated FDTD Performance,” Applied

Computational Electromagnetic Society (ACES)

Journal, vol. 25, no. 4, pp. 315-322, Apr. 2010.

M. J. Inman, A. Z. Elsherbeni, J. G. Maloney, and

B. N. Baker, “PracticalImplementation of a

CPML Absorbing Boundary for GPU Accelerated

FDTD Technique,” Applied Computational

Electromagnetic Society (ACES) Journal, vol. 23,

no. 1, pp. 16-22, Mar. 2008.

N. Takada, N. Masuda, T. Tanaka, Y. Abe, and T.

Ito, “A GPUImplementation of the 2-D Finite-

Difference Time-Domain Code using High Level

Shader Language,” Applied Compuational

Electromagnetic Society (ACES) Journal, vol. 23,

no. 4, pp. 309-316, Dec. 2008.

Intel Corporation, Intel Architecture Optimization

Reference Manual, Availab

http://www.intel.com/design/pentiumii/manuals/2

htm.

http://www.tecchannel.de/server/hardware/437111

/wechsel_an_der_spitze_intels_neue_core_prozess

oren/index9.html

http://software.intel.com/en-

us/articles/introduction-to-intel-advanced-vector-

extensions/

W. Yu, “ANovel Hardware Acceleration

Technique for High Performance Parallel FDTD

Method,” Microwave Technology &

Computational Electromagnetics (ICMTCE), 2011

IEEE International Conference on, pp. 441-444,

May 2011.

W. Yu, X. Yang, Y. Liu, et al., Advanced FDTD

Methods: Parallelization, Acceleration, and Engineering Applications, Artech House, Boston,

June 2011.

M. Yu, “Power-Handling Capability for RF

Filters,” IEEE Microwave Magazine, vol. 8, no. 5,

pp. 89-97, Oct. 2007.

J. M. Jin,The Finite Element Method in

Electromagnetics, New York: John Wiley & Sons,

Downloads

Published

2012-01-01

How to Cite

[1]
L. . Zhang, X. . Yang, and W. . Yu, “Enhanced Parallel FDTD Method Using SSE Instruction Sets”, ACES Journal, vol. 27, no. 1, pp. 1–8, Jan. 2012.

Issue

Section

General Submission