phone +7 (3412) 91 60 92

Archive of Issues


Russia Novosibirsk
Year
2018
Volume
28
Issue
3
Pages
419-426
<<
Section Computer science
Title The peculiarities of the parallel implementation of Particle-In-Cell method
Author(-s) Romanenko A.A.a, Snytnikov A.V.b
Affiliations Novosibirsk State Universitya, Institute of Computational Mathematics and Mathematical Geophysics, Siberian Branch of the Russian Academy of Sciencesb
Abstract Particle-In-Cell (PIC) method is widely used for plasma simulation and the GPUs appear to be the most efficient way to run this method. In this work we propose a technique that enables one to speed up one of the most time-consuming operations in the GPU implementation of the PIC method. The operation is particle reordering, or redistribution of particles between cells, which is performed after pushing. The reordering operation provides data locality which is the key performance issue of the PIC method. We propose to divide the reordering into two stages. First, gather the particles that are going to leave a particular cell into arrays, the number of arrays being equal to the number of neighbor cells (26 for 3D case). Second, each neighbor cell copies the particles from the necessary array to its own particle array. The second operation is done in 26 threads independently with no synchronization or waiting and involves no critical sections, semaphores, mutexes, atomic operations etc. It results in the more than 10 times reduction of the reordering time compared to the straightforward reordering algorithm.
Keywords optimization, GPU, simulation, PIC
UDC 519.684
MSC 68U20, 68N99
DOI 10.20537/vm180311
Received 23 May 2018
Language English
Citation Romanenko A.A., Snytnikov A.V. The peculiarities of the parallel implementation of Particle-In-Cell method, Vestnik Udmurtskogo Universiteta. Matematika. Mekhanika. Komp'yuternye Nauki, 2018, vol. 28, issue 3, pp. 419-426.
References
  1. Astrelin V.T., Burdakov A.V., Postupaev V.V. Generation of ion-acoustic waves and suppression of heat transport during plasma heating by an electron beam, Plasma Physics Reports, 1998, vol. 24, issue 5, pp. 414-425.
  2. Burau H., Widera R., Honig W., Juckeland G., Debus A., Kluge T., Schramm U., Cowan T.E., Sauerbrey R., Bussmann M. PIConGPU: a fully relativistic particle-in-cell code for a GPU cluster, IEEE Transactions on Plasma Science, 2010, vol. 38, issue 10, pp. 2831-2839. DOI: 10.1109/TPS.2010.2064310
  3. Rossi F., Londrillo P., Sgattoni A., Sinigardi S., Turchetti G. Towards robust algorithms for current deposition and dynamic load-balancing in a GPU particle in cell code, AIP Conference Proceedings, 2012, vol. 1507, issue 1, pp. 184-192. DOI: 10.1063/1.4773692
  4. Kong X., Huang M., Ren Ch., Decyk V. Particle-in-cell simulations with charge-conserving current deposition on graphic processing units, Journal of Computational Physics, 2011, vol. 230, issue 4, pp. 1676-1685. DOI: 10.1016/j.jcp.2010.11.032
  5. Rieke M., Trost T., Grauer R. Coupled Vlasov and two-fluid codes on GPUs, Journal of Computational Physics, 2015, vol. 283, pp. 436-452. DOI: 10.1016/j.jcp.2014.12.016
  6. Lotov K.V., Timofeev I.V., Mesyats E.A., Snytnikov A.V., Vshivkov V.A. Note on quantitatively correct simulations of the kinetic beam-plasma instability, Physics of Plasmas, 2015, vol. 22, issue 2, 024502. DOI: 10.1063/1.4907223
  7. Tskhakaya D., Schneider R. Optimization of PIC codes by improved memory management, Journal of Computational Physics, 2007, vol. 225, issue 1, pp. 829-839. DOI: 10.1016/j.jcp.2007.01.002
  8. Romanenko A.A., Snytnikov A.V., Timofeev I.V. 3D hybrid code for simulation of high-frequency electromagnetic radiation generation in turbulent plasma, Vestn. Novosib. Gos. Univ. Ser. Inf. Tekhnol., 2016, vol. 14, issue 3, pp. 81-90 (in Russian). https://elibrary.ru/item.asp?id=28098904
  9. Snytnikov A., Romanenko A. Parallel template implementation of Particle-in-Cell method for hybrid supercomputers, Bulletin of the Novosibirsk Computing Center. Series: Computer Science, 2014, vol. 36, pp. 79-88. https://elibrary.ru/item.asp?id=25467821
  10. Snytnikov A., Romanenko A. The advantage of the GPU-based supercomputer simulation of plasma phenomena, Bulletin of the Novosibirsk Computing Center. Series: Numerical Analysis, 2013, vol. 16, pp. 81-92. https://nccbulletin.ru/article/706
  11. Snytnikov A.V., Boronina M.A., Mesyats E.A., Romanenko A.A. A technology for the design of hybrid supercomputer simulation codes for relativistic particle electrodynamics, Proceedings of the 1st Russian Conference on Supercomputing, 2015, pp. 17-25. http://ceur-ws.org/Vol-1482/017.pdf
  12. Grigoryev Yu.N., Vshivkov V.A., Fedoruk M.P. Numerical “Particle-in-Cell” Methods, VSP, Utrecht-Boston, 2002.
Full text
<< Previous article