桌面计算机上利用格子Boltzmann方法的GPU计算

摘要/Abstract

摘要：

介绍了在桌面计算机上利用格子Boltzmann方法（LBM）与图形处理器（GPU）计算的发展背景，分析了LBM的标准形式及其天生并行特性的成因，介绍了所采用的CUDA编程模型及Kepler计算架构.为了验证桌面计算机上利用LBM的GPU计算的应用能力，对二维方柱绕流问题进行了数值模拟，并将模拟结果与有限体积法的计算结果进行对比.结果表明：对于方柱绕流问题，GPU计算的模拟计算效率约为CPU计算的3.4倍，桌面计算机上利用LBM的GPU计算具有一定的通用科学计算能力.

关键词: 格子Boltzmann方法, 图形处理器, 计算流体力学, 方柱绕流, 高性能计算

Abstract:

Development background of the combination of lattice Boltzmann method (LBM) and graphic processing unit (GPU) computing on a desktop computer was discussed. A detailed analysis was conducted of the standard LBM form and the reason of its natural parallelism. The CUDA programming model and the Kepler computing architecture used in the computing were expounded. In order to verify the application capacity of GPU computing of LBM on a desktop computer, numerical simulations of two-dimensional flow around a square cylinder were conducted, and simulation results with the finite volume method calculation results were compared. The results show that as for the problem of twodimensional flow around a square cylinder, computational efficiency of GPU computing is approximately 3.4 times of central processing unit (CPU) computing, which proved the computing capacity of GPU computing of LBM on desktop computers to universal scientific computing.

Key words: lattice Boltzmann method (LBM), graphic processing unit (GPU), computational fluid dynamics, flow around a square cylinder, high performance computing

中图分类号:

O 357.1

刘强，谢伟，邱辽原，解学参. 桌面计算机上利用格子Boltzmann方法的GPU计算[J]. 上海交通大学学报（自然版）, 2014, 48(09): 1329-1333.

LIU Qiang,XIE Wei,QIU Liaoyuan,XIE Xueshen. Graphic Processing Unit Computing of Lattice Boltzmann Method on a Desktop Computer[J]. Journal of Shanghai Jiaotong University, 2014, 48(09): 1329-1333.

参考文献

［1］Bernaschi M, Fatica M, Melchionna S, et al. A flexible highperformance lattice Boltzmann GPU code for the simulations of fluid flows in complex geometries ［J］. Concurrency and Computation: Practice and Experience, 2010, 22 (1): 114.

［2］Tanno I, Hashimoto T, Yasuda T, et al. Simulation of turbulent flow by lattice Boltzmann method and conventional method on a GPU ［J］. Computers and Fluids, 2013, 80 (10): 453458.

［3］Obrecht C, Kuznik F, Tourancheau B, et al. A new approach to the lattice Boltzmann method for graphics processing units ［J］. Computers and Mathematics with Applications, 2011, 61 (12): 36283638.

［4］Habich J, Feichtinger C, Kstler H, et al. Performance engineering for the lattice Boltzmann method on GPGPUs: Architectural requirements and performance results ［J］. Computers and Fluids, 2013, 80 (10): 276282.

［5］Xian W, Takayuki A. MultiGPU performance of incompressible flow computation by lattice Boltzmann method on GPU cluster ［J］. Parallel Computing, 2011, 37 (9): 521535.

［6］Schnherr M, Kucher K, Geier M, et al. Multithread implementations of the lattice Boltzmann method on nonuniform grids for CPUs and GPUs ［J］. Computers and Mathematics with Applications, 2011, 61 (12): 37303743.

［7］何雅玲, 王勇, 李庆. 格子Boltzmann方法的理论及应用［M］. 北京: 科技出版社, 2009: 117.

［8］Chen S, Doolen G. Lattice Boltzmann method for fluid flows ［J］.Annual Review of Fluid Mechanics, 1998, 30 (1): 329364.

［9］Aidun C, Clausen J. LatticeBoltzmann method for complex flows ［J］. Annual Review of Fluid Mechanics, 2010, 42 (1): 439472.

［10］Habich J, Zeiser T, Hager G, et al. Performance analysis and optimization strategies for a D3Q19 lattice Boltzmann kernel on nVIDIA GPUs using CUDA ［J］. Advances in Engineering Software, 2011, 42 (5): 266272.

［11］Lammers P, Beronov K, Volkert R, et al. Lattice BGK direct numerical simulation of fully developed turbulence in incompressible plane channel flow ［J］. Computers and Fluids, 2006, 35 (10): 11371153.

［12］Breuer M, Bernsdorf J, Zeiser T, et al. Accurate computations of the laminar flow past a square cylinder based on two different methods: LatticeBoltzmann and finitevolume ［J］. International Journal of Heat and Fluid Flow, 2000, 21 (2): 186196.

[1]	王志伟, 何炎平, 李铭志, 仇明, 黄超, 刘亚东. 基于计算流体力学的90° 弯管气液两相流数值模拟及流型演化[J]. 上海交通大学学报, 2022, 56(9): 1159-1167.
[2]	高昌昊, 宋文萍, 韩少强, 路宽, 王跃, 叶坤. 空空导弹过失速重新定向技术研究[J]. 空天防御, 2022, 5(3): 17-26.
[3]	陈志鑫, 汪怡平, 杨亚锋, 苏建军, 杨斌. 不同送风方式下大客车内飞沫传播特性研究[J]. 上海交通大学学报, 2022, 56(11): 1532-1540.
[4]	张宇, 王晓亮. 基于径向点插值方法的柔性螺旋桨气动弹性模拟[J]. 上海交通大学学报, 2020, 54(9): 924-934.
[5]	王瑞, 肖瑶, 顾汉洋, 叶亚楠. 螺旋管内单相流动周向非均匀传热现象的数值模拟[J]. 上海交通大学学报, 2020, 54(7): 688-696.
[6]	. 半潜式钻井平台风载特征及影响因素分析[J]. 海洋工程装备与技术, 2019, 6(3): 548-.
[7]	顾娟，黄荣宗，刘振宇，吴慧英. 不同热边界条件下微通道内气体的流动与传热特性[J]. 上海交通大学学报（自然版）, 2018, 52(9): 1038-1043.
[8]	郁程，董小倩，杨晨俊. 侧推器体积力模型及其应用[J]. 上海交通大学学报（自然版）, 2018, 52(3): 291-296.
[9]	李懿霖,宋保维. 空化器直径对超空泡航行器空泡性能的影响[J]. 上海交通大学学报（自然版）, 2017, 51(12): 1488-1492.
[10]	米百刚，詹浩. 先进飞行器动导数数值模拟新方法[J]. 上海交通大学学报（自然版）, 2016, 50(04): 619-624.
[11]	刘承江1，王永生1，古成中2. 船-泵相互作用对喷水推进器推进性能的影响[J]. 上海交通大学学报（自然版）, 2016, 50(01): 91-97.
[12]	刘晗a，马宁a,b*，邵闯a，顾解忡a,b. 限宽水域中船舶平面运动机构试验及水动力导数数值模拟[J]. 上海交通大学学报（自然版）, 2016, 50(01): 115-122.
[13]	周振龙，朱锡，张帅. 螺旋桨CFD不确定度及叶形对桨叶变形的影响[J]. 上海交通大学学报（自然版）, 2014, 48(1): 74-80.
[14]	龚帅,郑平. 液滴在非均匀润湿表面上的动力行为[J]. 上海交通大学学报（自然版）, 2014, 48(02): 239-244.
[15]	严立，戴欣怡，陈佳洛，王平阳，欧阳华. 基于计算统一设备架物Fortran的直接模拟蒙特卡洛方法并行优化[J]. 上海交通大学学报（自然版）, 2013, 47(08): 1198-1204.