Your browser does not support JavaScript!
http://iet.metastore.ingenta.com
1887

access icon free Efficient strategy for parallelisation of multilevel fast multipole algorithm using CUDA

© The Institution of Engineering and Technology
Received 17/07/2018, Accepted 08/04/2019, Revised 27/02/2019, Published 10/04/2019

The multilevel fast multipole algorithm is a popular technique that enables the efficient solution of the method of moments (MoM) matrix equations. In this work, the authors address the adaptation of this method to the compute unified device architecture (CUDA), a relatively new computing infrastructure provided by NVIDIA, and the authors take into account some of the limitations that appear when the geometry under analysis becomes too large to fit into the memory of graphics processing units.

Inspec keywords: method of moments; electromagnetic wave scattering; computer graphic equipment; parallel architectures; coprocessors; graphics processing units; computational electromagnetics

Other keywords: CUDA; moments matrix equations; efficient strategy; efficient solution; multilevel fast multipole algorithm; relatively new computing infrastructure; compute unified device architecture; popular technique

Subjects: Microprocessors and microcomputers; Parallel architecture; Numerical approximation and analysis; Electromagnetic wave propagation; Microprocessor chips

References

    1. 1)
      • 14. Lezar, E., Davidson, D.B.: ‘GPU acceleration of method of moments matrix assembly using Rao-Wilton-Glisson basis functions’. 2010 Int. Conf. Electronics and Information Engineering (ICEIE 2010), Kyoto, Japan, 2010.
    2. 2)
      • 9. Lezar, E., Davidson, D.B.: ‘GPU-accelerated method of moments by example: monostatic scattering’, IEEE Trans. Antennas Propag., 2010, 52, (6), pp. 120–135.
    3. 3)
      • 25. Velamparambil, S., Chew, W.C., Song, J.: ‘10 million unknowns: is it that big?’, IEEE Antennas Propag. Mag., 2003, 45, (2), pp. 43–58.
    4. 4)
      • 19. Guan, J., Yan, S., Jin, J.-M.: ‘An OpenMP-CUDA implementation of multilevel fast multipole algorithm for electromagnetic simulation on multi-GPU computing systems’, IEEE Trans. Antennas Propag., 2013, 61, (7), pp. 3607–3616.
    5. 5)
      • 27. Nickolls, J., Buck, I., Garland, M., et al: ‘Scalable parallel programming’, ACM. Queue., 2008, 6, (2), pp. 40–53.
    6. 6)
      • 7. Ludick, D.J., Davidson, D.B.: ‘Investigating efficient parallelization techniques for the characteristic basis function method (CBFM)’. 2009 Int. Conf. Electromagnetics in Advanced Applications (ICEAA 09), Torino, Italy, 2009.
    7. 7)
      • 12. Gurel, L., Ergul, O.: ‘Hierarchical parallelization of the multilevel fast multipole algorithm (MLFMA)’, Proc. IEEE, 2013, 101, (2), pp. 332–341.
    8. 8)
      • 23. Cátedra, M.F., Rivas, F., Valle, L.: ‘Combining the moment method with geometrical modelling by NURBS surfaces and Bezier patches’, IEEE Trans. Antennas Propag., 1994, 42, (3), pp. 373–381.
    9. 9)
      • 18. Guan, J., Yan, S., Ming Jin, J.: ‘An accurate and efficient finite element-boundary integral method with GPU acceleration for 3-D electromagnetic analysis’, IEEE Trans. Antennas Propag., 2014, 62, (12), pp. 6325–6336.
    10. 10)
      • 15. Zoric, D.P., Olcan, D.I., Kolundzija, B.M.: ‘Solving electrically large EM problems by using out-of-core solver accelerated with multiple graphical processing units’. 2011 IEEE Int. Symp. Antennas and Propagation and USNC/URSI National Radio Science Meeting (APSURSI 2011), Rome, Italy, 2011.
    11. 11)
      • 1. Harrington, R.F.: ‘Field computation by moment methods’ (McMillan, New York, 1968).
    12. 12)
      • 22. ]Valle, L., Rivas, F., Cátedra, M.F.: ‘A moment method approach using frequency independent parametric meshes’, IEEE Trans. Antennas Propag., 1997, 45, (10), pp. 1567–1568.
    13. 13)
      • 4. Chew, W.C., Jin, J., Michielssen, E., Song, J. (Eds.): ‘Fast and efficient algorithms in computational electromagnetics’ (Artech House, Norwood, MA USA, 2001).
    14. 14)
      • 8. García, E., Lozano, L., Algar, M.J., et al: ‘‘A study of the efficiency of the parallelization of a high frequency electromagnetic approach for the computation of radiation and scattering considering multiple bounces’, Comput. Phys. Commun., 2013, 184, (1), pp. 45–50.
    15. 15)
      • 13. Pan, X.M., Pi, W.C., Yang, M.L., et al: ‘Solving problems with over one billion unknowns by the MLFMA’, IEEE Trans. Antennas Propag., 2012, 60, (5), pp. 2571–2574.
    16. 16)
      • 16. Peng, S., Nie, Z.: ‘Acceleration of the method of moments calculations by using graphics processing units’, IEEE Trans. Antennas Propag., 2008, 56, (7), pp. 2130–2133.
    17. 17)
      • 10. Zoric, D.P., Olcan, D.I., Kolundzija, B.M.: ‘GPU accelerated computation of radar cross sections with multiple excitations’. 2013 European Conf. Antennas and Propagation (EUCAP 2013), 2013.
    18. 18)
      • 11. Michiels, B., Fostier, J., Bogaert, I., et al: ‘Full-wave simulations of electromagnetic scattering problems with billions of unknowns’, IEEE Trans. Antennas Propag., 2015, 63, (2), pp. 796–799.
    19. 19)
      • 20. Xu, K., Ding, D.Z., Fan, Z.H., et al: ‘Multilevel fast multipole algorithm enhanced by GPU parallel technique for electromagnetic scattering problems’, Microw. Opt. Technol. Lett., 2010, 52, (3), pp. 502–507.
    20. 20)
      • 17. Topa, T., Karwowski, A., Noga, A.: ‘Using GPU with CUDA to accelerate MoM-based electromagnetic simulation of wire-grid models’, IEEE Antennas Wirel. Propag. Lett., 2011, 10, pp. 342–345.
    21. 21)
      • 24. Press, W.H., Flannery, B.P., Teukolsky, S.A., et al: ‘Numerical recipes in FORTRAN: the art of scientific computing’ (Cambridge University Press, Cambridge, 1992, 2nd edn.).
    22. 22)
      • 5. Ding, D.Z., Fan, Z.H., Tao, S.F., et al: ‘Complex source beam method for EM scattering from PEC objects’, IEEE Antennas Wirel. Propag. Lett., 2015, 14, pp. 346–349.
    23. 23)
      • 3. Engheta, N., Murphy, W.D., Rokhlin, V., et al: ‘The fast multipole method (FMM) for electromagnetic scattering problems’, IEEE Trans. Antennas Propag., 1992, 40, (6), pp. 634–641.
    24. 24)
      • 28. NVIDIA Corporation: ‘CUDA programming guide’ (NVIDIA, Santa Clara, CA, 2013). Available at http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html.
    25. 25)
      • 6. Ding, D.Z., Chen, G.S., Chen, R., et al: ‘An efficient algorithm for surface integral equation based on meshfree scheme’, IEEE Antennas Wirel. Propag. Lett., 2014, 13, pp. 1541–1544.
    26. 26)
      • 26. González, I., García, E., Sáez De Adana, F., et al: ‘Monurbs: a parallelized fast multipole multilevel code for analysing complex bodies modelled by NURBS surfaces’, Appl. Comput. Electromagn. Soc. J., 2008, 23, (2), pp. 134–142.
    27. 27)
      • 21. Rivas, F., Valle, L., Cátedra, M.F.: ‘A moment method formulation for the analysis of wire antennas attached to arbitrary conducting bodies defined by parametric surfaces’, Appl. Comput. Electromagn. Soc. J., 1996, 11, (2), pp. 32–39.
    28. 28)
      • 2. Prakash, V.V.S., Mittra, R.: ‘Characteristic basis function method: a new technique for efficient solution of method of moments matrix equation’, Microw. Opt. Technol. Lett., 2003, 36, (2), pp. 95–100.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-map.2018.5568
Loading

Related content

content/journals/10.1049/iet-map.2018.5568
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
Print this page
This is a required field
Please enter a valid email address