LBM伪势MRT三维模型GPU并行计算的性能优化

格子Boltzmann方法伪势模型算法中的格点间计算未完全局部化，因此在并行计算时需要更多次的全局内存读写、使用更多数量的寄存器和线程同步操作，从而导致GPU并行计算效率下降.本文针对伪势模型并行计算的局限性，基于三维十五速格子结构的多松弛时间伪势模型，以气液相分离为算例，通过合并访问的方式提高全局内存的读写效率；并提出一种"定向转移"算法，提高格子边界格点获取邻居格点数据的效率；最后探索不同资源分配中各种因素对计算效率的影响，总结最优资源分配的方法.

Performance Optimization of 3D Pseudopotential Multi-Relaxation-Time Lattice Boltzmann Model on GPU

Pseudopotential model of lattice Boltzmann method is partially non-local for pseudopotential calculation with coupling of lattices, which leads to synchronization of threads in parallel implementation process. Besides, it uses a large number of registers and much time of data access operations when access global memory in calculation process. They lead to low computational efficiency. In this paper, a multi-relaxation-time(MRT) 3D pseudopotential model with D3Q15 lattice is adopted as an example to investigate performance of parallel computing based on GPU. To address limitation of parallel computing of pseudo-potential model, efficiency of reading and writing of global memory is improved by using merge access method. To improve efficiency of grids retrieving data which are in boundary of lattice, a "Directional Transfer" algorithm is proposed. The role of computing resource configuration is investigated with different sizes of block, and optimal resource configuration scheme is obtained.