A GPU parallel staircase finite difference mesh generation algorithm based on the ray casting method

被引:0
作者
Li P. [1 ]
Ma T. [1 ]
Xu X. [1 ]
Ma T. [1 ]
机构
[1] School of Mechatronical Engineering, Beijing Institute of Technology, Beijing
来源
Baozha Yu Chongji/Explosion and Shock Waves | 2020年 / 40卷 / 02期
关键词
Data transmission strategy; GPU parallel computing; Mesh generation; Ray casting method;
D O I
10.11883/bzycj-2019-0344
中图分类号
学科分类号
摘要
Three-dimensional large-scale finite difference mesh generation technology is the basis of three-dimensional finite difference computation, and the efficiency of mesh generation is a research hotspot of three-dimensional finite difference mesh generation. The traditional staircase finite difference mesh generation algorithm mainly includes ray casting algorithm and slicing algorithm. Based on the traditional serial ray casting algorithm, a parallel staircase finite difference mesh generation algorithm based on GPU (graphic processing unit) is proposed in this paper. Parallel algorithm uses batch-based data transmission strategy, which makes the scale of mesh generation independent of GPU memory size, and balances the relationship between data transmission efficiency and mesh generation scale. In order to reduce the time consumption of data transmission between the host memory and the device memory, the parallel algorithm proposed in this paper can generate ray starting coordinates independently within GPU threads, which further improves the execution efficiency and parallelization degree of the parallel algorithm. The comparison of numerical experiments shows that the efficiency of parallel algorithm is much higher than that of traditional ray casting algorithm. Finally, an example of finite difference calculation shows that the parallel algorithm can meet the requirement of large-scale numerical simulation of complex models. © 2020, Editorial Staff of EXPLOSION AND SHOCK WAVES. All right reserved.
引用
收藏
相关论文
共 20 条
[1]  
Ma T.B., Ren H.L., Li J., Et al., Large scale high precision computation for explosion and impact problems, Chinese Journal of Theoretical and Applied Mechanics, 48, 3, pp. 599-608, (2016)
[2]  
Ning J.G., Yuan X.P., Ma T.B., Et al., Positivity-preserving moving mesh scheme for two-step reaction model in two dimensions, Computers and Fluids, 123, pp. 72-86, (2015)
[3]  
Wang X., Ma T.B., Ning J.G., A pseudo arc-length method for numerical simulation of shock waves, Chinese Physics Letters, 31, 3, (2014)
[4]  
Chen L.W., Zhang H., Wang X.G., Three-dimensional numerical simulation of multi-material explosive field in water, Acta Armamentarii, pp. 1-4, (2009)
[5]  
Zhang J., Zhao N., Ren D.F., Et al., Application of the level set method on adaptive Cartesian grids, Explosion and Shock Waves, 28, 5, pp. 438-442, (2008)
[6]  
Xiao H.S., Liu G., Chen Z.B., Et al., The adaptive Cartesian grid generation method based on STL file, Acta Aerodynamica Sinica, 24, 1, pp. 120-124, (2006)
[7]  
Pandey P.M., Reddy N.V., Dhande S.G., Slicing procedures in layered manufacturing: a review, Rapid Prototyping Journal, 9, 5, pp. 274-288, (2003)
[8]  
Zhao J.B., Liu W.J., Recent progress in slicing algorithm of rapid prototyping technology, Computer Integrated Manufacturing Systems, 15, 2, pp. 209-221, (2009)
[9]  
Fei G.L., Ma T.B., Hao L., Large-scale high performance computation on 3D explosion and shock problems, Applied Mathematics and Mechanics, 32, 3, pp. 375-382, (2011)
[10]  
Macgillivray J.T., Trillion cell CAD-based Cartesian mesh generator for the finite-difference time-domain method on a single-processor 4-GB workstation, IEEE Transactions on Antennas and Propagation, 56, 8, pp. 2187-2190, (2008)