Keywords
Yee cell
data parallel model
shared memory system
task
speedup
Amdahl’s law
How to Cite
Abstract
The finite-difference time-domain (FDTD) is a numerical analysis technique used for solving computational electrodynamic problems. The nature of the FDTD method is that simulation of big and complicated electromagnetic field problems requires a vast amount of computer operational memory and runtime. Parallel-processing techniques have been broadly applied to FDTD to accelerate the simulations.
The parallelism of the FDTD algorithm is based on a fact that the computational domain can be divided into parts (sub-domains), and each processor in a parallel system deals with one or several sub-domains. The FDTD algorithm belongs to data parallelism model and can be effectively implemented on shared memory system architecture.
The parallel FDTD method was implemented using TPL library. The Task Parallel Library (TPL) is a library for .NET that makes easy to parallelize the program using the advantages of .NET Framework. The speedup metrics of parallel FDTD algorithm were calculated and compared with Amdahl’s estimated speedup.
References
Yee, K. S., Numerical solution of initial boundary value problems involving Maxwell’s equations in isotropic media, IEEE Transactions on Antennas and Propagation, Vol. 14, No. 3, 1966, pp. 302–307.
Taflove, A. and Hagness, S., Computational Electrodynamics: The Finite-Difference Time Domain Method, Artech House, Norwood, 2006.
He, Z. L., Huang, K., Zhang, Y., Yan, Y., and Liang, C. H., Study on High Performance of MPI-Based Parallel FDTD fromWorkStation to Super Computer Platform, International Journal of Antennas and Propagation, Vol. 2012, 2012, Article ID 659509, 7 pages.
Guo, X. M., Guo, Q. X., Zhao,W., and Yu,W. H., Parallel FDTD simulation using NUMA acceleration technique, Progress In Electromagnetics Research Letters, Vol. 28, 2012, pp. 1–8.
Sen, R., Developing Parallel Programs, Advances in Computer Science: an International Journal, Vol. 1, No. 1, 2012, pp. 18–27, Available at: http://www.acsij.org/acsij/article/view/321.
Barney, B., Introduction to Parallel Computing, Lawrence Livermore National Laboratory, 2015, Available at: https://computing.llnl.gov/tutorials/parallel_comp/.
Demir, V. and Elsherbeni, A. Z., Compute Unified Device Architecture (CUDA) based Finite-Di erence Time-Domain (FDTD) implementation, ACES Journal, Vol. 25, No. 4, 2010, pp. 303–314.
Ujaldon, M., Using GPUs for accelerating electromagnetic simulations, ACES Journal, Vol. 25, No. 4, 2010, pp. 294–302.
Task Parallel Library (TPL), 2012, Available at: https://msdn.microsoft.com/en-us/library/dd460717(v=vs.110).aspx.
Leijen, D., Schulte, W., and Burckhardt, S., The Design of a Task Parallel Library, In: Proceedings of the 24th ACM SIGPLAN Conference on Object Oriented Programming Systems Languages and Applications, OOPSLA ’09, ACM, 2009, pp. 227–242.
Vištica, M., Haseljic, H., Maksumi´c, A., and Nosovic, N., Comparison of speedups for computing using .NET TPL and OpenMP parallelization techonologies, In: X International Symposium on Telecommunications (BIHTEL), 2014.
Amdahl, G. M., Validity of the Single Processor Approach to Achieving Large Scale Computing Capabilities, In: Proceedings of the April 18-20, 1967, Spring Joint Computer Conference, AFIPS ’67 (Spring), ACM, 1967, pp. 483–485.
Zhang, K. D., Shao,W., andWang, B. Z., Parallelized ADI-FDTD Algorithm for Attenuation Constant Extraction by Using OpenMP Library, In: International Conference on Microwave and Millimeter Wave Technology, IEEE, 2010, pp. 786–788.