Parallel Overlapping Mechanism Between Communication and Computation of the Lattice Boltzmann Method

详细信息    查看全文

• 关键词：Lattice Moltzmann Method ; Single ; Relaxation ; Time ; Overlapping communication and computation ; Hybrid model ; Parallel model
• 刊名：Lecture Notes in Computer Science
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：9576
• 期：1
• 页码：196-203
• 全文大小：455 KB
• 参考文献：1.Succi, S.: The lattice Boltzmann equation for fluid dynamics and beyond. Oxford Science Publications, Oxford (2001)MATH
  2.Aidun, C.K., Clausen, J.R.: Lattice-Boltzmann method for complex flows. Ann. Rev. Fluid Mech. 42, 439–472 (2010)MathSciNet CrossRef MATH
  3.Chen, S., Doolen, G.D.: Lattice Boltzmann method for fluid flows. Ann. Rev. Fluid Mech. 30, 329–364 (1998)MathSciNet CrossRef
  4.Qian, Y., d’Humières, D., Lallemand, P.: Lattice BGK models for Navier-Stokes equation. Europhys. Lett. 17, 479–484 (1992)CrossRef MATH
  5.Filippovaa, O., Succib, S., Mazzoccoc, F., et al.: Multiscale lattice Boltzmann schemes with turbulence modeling. J. Comput. Phys. 170, 812–829 (2001)CrossRef
  6.Hou, S., Sterling, J., Chen, S., Doolen, G.D.: A lattice Boltzmann subgrid model for high Reynolds number flows fields. Fields Inst, Commun. 6, 151–166 (1996)MathSciNet MATH
  7.Dazhi, Y., Mei, R., Shyy, W.: A multi-block lattice Boltzmann method for viscous fluid flows. Int. J. Numer. Methods Fluids 39, 99–120 (2002)CrossRef MATH
  8.Jelinek, B., Eshraghi, M., Felicelli, S., et al.: Large-scale parallel lattice Boltzmann-cellular automaton model for two-dimensional dendritic growth. Comput. Phys. Commun. 185, 939–947 (2014)CrossRef
  9.Liu, Z., Song, A., Lei, X., et al.: A high scalable hybrid MPI/OpenMP parallel model of Multiple-Relaxation-Time lattice Boltzmann method. J. Comput. Inf. Syst. 23, 10147–10157 (2014)
  10.Vidal, D., Roy, R., Betrand, F.: A parallel workload balanced and memory efficient lattice-Boltzmann algorithm with single unit BGK relaxation time for laminar Newtonian flows. Comput. Fluids 39, 1411–1423 (2010)MathSciNet CrossRef MATH
  11.Schepke, C., Maillard, N., Philippe, O.A.: Navaux, Parallel lattice Boltzmann method with blocked partitioning. Int. J. Parallel Program. 37, 593–611 (2009)CrossRef MATH
  12.Ye, Y., Kenli, Y., Wang, Y., Deng, T.: Parallel computation of entropic lattice Boltzmann method on hybrid CPU-GPU accelerated system. Comput. Fluids 110, 114–121 (2015)MathSciNet CrossRef
  13.Wang, Z., Zhao, Y., Sawchuck, A.P., et al.: GPU acceleration of volumetric lattice Boltzmann method for patient-specific computational hemodynamics. Comput. Fluids 115, 192–200 (2015)MathSciNet CrossRef
  14.Ghia, U., Ghia, K.N., Shin, C.T.: High-Re solutions for incompressible flow using the Navier-Stokes equations and a multigrid method. J. Comput. Phys. 48, 387–411 (1982)CrossRef MATH
  
• 作者单位：Zhixiang Liu (18) (19)
  Yong Fang (18)
  Anping Song (20)
  Lei Xu (20)
  Xiaowei Wang (19) (20)
  Liping Zhou (19) (20)
  Wu Zhang (19) (20)
  
  18. School of Communication and Information Engineering, Shanghai University, Shanghai, 200444, China
  19. High Performance Computing Center, Shanghai University, Shanghai, 200444, China
  20. School of Computer Engineering and Science, Shanghai University, Shanghai, 200444, China
  
• 丛书名：High Performance Computing and Applications
• ISBN：978-3-319-32557-6
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349
• 卷排序：9576

文摘

The lattice Boltzmann Method (LBM), different from classical numerical methods of continuum mechanics, is derived from molecular dynamics. The LBM has the following main advantages: including a simple algorithm, the direct solver for pressure, easy treatment of complicated boundary conditions and particularly parallel suitability. The most common models include the Single-Relaxation-Time (SRT) and Multiple-Relaxation-Time (MRT) collision models. In a conventional parallel computing model of LBM, communication and computing are performed individually. When the communication is performed, the computing is waiting in MPI processes. This will waste some waiting time. Therefore, the communication and computing overlapping parallel model was proposed. By the architecture of “Ziqiang 4000” supercomputer at Shanghai University, the hybrid MPI and OpenMP parallel model is proposed. The numerical results show that the presented model has better computational efficiency.