(43-4) 20 * << * >> * Russian * English * Content * All Issues

Design features of block algorithms of FDTD-method implemented on a GPU using MATLAB

N.D. Morunov1, D.L. Golovashkin1,2

1 Samara National Research University,
Moskovskoye Shosse 34, 443086, Samara, Russia,

2 IPSI RAS – Branch of the FSRC “Crystallography and Photonics” RAS,
Molodogvardeyskaya 151, 443001, Samara, Russia

 PDF, 799 kB

DOI: 10.18287/2412-6179-2019-43-4-671-676

Pages: 671-676.

Full text of article: Russian language.

Abstract:
The paper is devoted to the investigation of the implementation features of a block algorithm for the FDTD-method on GPU. The block algorithm in general and in the context of the FDTD-method in particular is discussed. The main attention is paid to specifics of determining the shape and volume of blocks, which stem from the use of MATLAB and its Parallel Computing Toolbox. The practical efficiency of the proposed techniques is shown. Possible applications and prospects are discussed.

Keywords:
FDTD-method, block algorithms, computational speed-up

Citation:
Morunov ND, Golovashkin DL. Design features of block algorithms of FDTD-method implemented on a GPU using MATLAB. Computer Optics 2019; 43(4): 671-676. DOI: 10.18287/2412-6179-2019-43-4-671-676.

References:
  1. Yee KS. Numerical solution of initial boundary value problems involving Maxwell's equations in isotropic media. IEEE Transactions on Antennas and Propagation 1966; AP-14: 302-307. DOI: 10.1109/TAP.1966.1138693.
  2. Kotlyar VV, Stafeev SS, Nalimov AG. Sharp focusing of laser light with micro-optics [In Russian]. Samara: "Novaya Tehnika" Publisher; 2018. ISBN: 978-5-88940-148-3.
  3. Krakiwsky SE, Turner LE, Okoniewski MM. Graphics processor unit (GPU) acceleration of finite-difference time-domain (FDTD) algorithm. Proc ISCAS'04 2004: 1033-1036. DOI: 10.1109/ISCAS.2004.1329513.
  4. Taflove A, Hagness S. Computational electrodynamics: The finite-difference time-domain method. 3th ed. Boston: Arthech House Publishers, 2005. ISBN: 978-1-58053-832-9.
  5. XStream® GPU Acceleration. Source: <https://www.remcom.com/xf-xstream>.
  6. Wahl P, LyGagnon DS, Debaes Ch, Miller DAB, Thienpont H. B-CALM: An open-source GPU-based 3D-FDTD with multi-pole dispersion for plasmonics. Opt Quant Electron 2012; 44(3): 285-290. DOI: 10.1007/s11082-012-9558-z.
  7. Elsherbeni AZ, Demir V. The finite-difference time-domain method for electromagnetics with MATLAB simulations. Ralrigh, NC: SciTech Publishing Inc; 2009. ISBN: 978-1-891121-71-5.
  8. BrookGPU. Source: <http://graphics.stanford.edu/projects/brookgpu>.
  9. Dinier JE, Elsherbeni AZ. FDTD Acceleration using MATLAB parallel computing toolbox and GPU. Applied Computational Electromagnetics Society Journal 2017; 32(4): 283-288.
  10. Zakirov AV, Levchenko VD, Perepelkina AYu, Zempo Y. DiamondTorre algorithm and high-performance implementation of the FDTD-method for supercomputers with graphics accelerators [In Russian]. Proc "Supercomputer days in Russia '16" 2016: 80-94.
  11. Golub GH, Van Loan ChF. Matrix Computations. 3rd ed. Baltomore, London: Johns Hopkins University Press; 1996. ISBN: 978-0-8018-5414-9.
  12. Demmel J. Applied numerical linear algebra. Philadelphia: SIAM; 1997. ISBN: 978-0-89871-389-3.
  13. Orozco D, Guang G. Mapping the FDTD application to many-core chip architectures. Proc ICPP '09 2009: 309-316. DOI: 10.1109/ICPP.2009.44.
  14. Minami T, Hibino M, Hiraishi T, Iwashita T, Nakashima H. Automatic parameter tuning of three-dimensional tiled FDTD kernel. In Book: Daydé M, Marques O, Nakajima K, eds. High performance computing for computational science (VECPAR 2014) 2014: 284-297. DOI: 10.1007/978-3-319-17353-5_24.
  15. Valkovskii V. Parallel execution cycles. Method of the pyramids [In Russian]. Cybernetics 1983; 5: 51-55.


© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: ko@smr.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20