Abstract
The Finite-Difference in Time-Domain (FDTD) method is widely used in many applications requiring to solve Maxwell’s Equations. Since simulations with large spaces, or long non-sinusoidal waveforms, imply high computational floating-point performance, it is of practical interest to take advantage of current and emergent multicore architectures, namely Graphics Processing Units (GPUs) (Pratas, et al.: Fine-grain parallelism using multi-core, cell/BE, and GPU systems: accelerating the phylogenetic likelihood function. In: International Conference on Parallel Processing, 2009 (ICPP’09), pp. 9–17. IEEE, Piscataway, 2009). The objective of the proposed chapter is to exploit data parallelism to efficiently compute the FDTD algorithm on multi-processors. Compute Unified Device Architecture (CUDA) and Open Computing Language (OpenCL) implementations of the parallel FDTD algorithm are presented and its relative performance evaluated. Source codes of the implementations for both frameworks is provided and comparison of results obtained for different GPUs, considering performance and scalability, is performed.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Notice that in CUDA the definition of events refers to timing features.
References
Acosta, A., Corujo, R., Blanco, V., Almeida, F.: Dynamic load balancing on heterogeneous multicore/multiGPU systems. In: 2010 International Conference on High Performance Computing and Simulation (HPCS), pp. 467–476 (2010). doi:10.1109/HPCS.2010.5547097
Kuan, L., Tomas, P., Sousa, L.: A comparison of computing architectures and parallelization frameworks based on a two-dimensional FDTD. In: 2013 International Conference on High Performance Computing and Simulation (HPCS), pp. 339–346 (2013). doi:10.1109/HPCSim.2013.6641436
Nvidia: http://docs.nvidia.com/cuda/pdf/Kepler_Tuning_Guide.pdf. 7 Feb 2014
Nvidia: http://www.nvidia.com/content/PDF/kepler/NVIDIA-Kepler-GK110-Architecture-Whitepaper.pdf. 7 Feb 2014
Pratas, F., Trancoso, P., Stamatakis, A., Sousa, L.: Fine-grain parallelism using multi-core, cell/BE, and GPU systems: accelerating the phylogenetic likelihood function. In: International Conference on Parallel Processing, 2009 (ICPP’09), pp. 9–17. IEEE, Piscataway (2009)
Shirahata, K., Sato, H., Matsuoka, S.: Hybrid map task scheduling for GPU-based heterogeneous clusters. In: 2010 IEEE Second International Conference on Cloud Computing Technology and Science (CloudCom), pp. 733–740 (2010). doi:10.1109/CloudCom.2010.55
Taflove, A., Hagness, S.C.: Computational electromagnetics: The Finite-Difference Time-Domain Method, Third Edition. Artech House, (2005)
Taflove, A., Brodwin, M.: Numerical solution of steady-state electromagnetic scattering problems using the time-dependent Maxwell’s equations. IEEE Trans. Microw. Theory Tech. 23(8), 623–630 (1975). doi:10.1109/TMTT.1975.1128640
Wittenbrink, C.M., Kilgariff, E., Prabhu, A.: Fermi GF100 GPU architecture. IEEE Micro 31(2), 50–59 (2011)
Yee, K., Chen, J.: The finite-difference time-domain (FDTD) and the finite-volume time-domain (FVTD) methods in solving Maxwell’s equations. IEEE Trans. Antennas Propag. 45(3), 354–363 (1997). doi:10.1109/8.558651
Zanjani, M., Akbari, A., Mirzaei, H., Shirdel, N., Gockenbach, E., Borsi, H.: Investigating partial discharge UHF electromagnetic waves propagation in transformers using FDTD technique and 3D simulation. In: 2012 International Conference on Condition Monitoring and Diagnosis (CMD), pp. 497–500 (2012). doi:10.1109/CMD.2012.6416187
Zhong, Z., Rychkov, V., Lastovetsky, A.: Data partitioning on heterogeneous multicore and multi-GPU systems using functional performance models of data-parallel applications. In: 2012 IEEE International Conference on Cluster Computing (CLUSTER), pp. 191–199. IEEE, Piscataway (2012)
Acknowledgements
The work presented herein was partially supported by national funds through Fundação para a Ciência e a Tecnologia (FCT) under projects Threads (ref. PTDC/ EEA-ELC/117329/2010), P2HCS (ref. PTDC/EEI-ELC/3152/2012) and PEst-OE/ EEI/LA0021/2013, and also with the Ph.D. grant with reference number SFRH/BD/ 65636/2009.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Kuan, L., Tomás, P., Sousa, L. (2014). Finite-Difference in Time-Domain Scalable Implementations on CUDA and OpenCL. In: Kindratenko, V. (eds) Numerical Computations with GPUs. Springer, Cham. https://doi.org/10.1007/978-3-319-06548-9_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-06548-9_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-06547-2
Online ISBN: 978-3-319-06548-9
eBook Packages: Computer ScienceComputer Science (R0)