arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.08262math.NAcs.NAphysics.comp-ph
收藏

用于双曲玻尔兹曼输运的约化多重网格的粗化与并行化

Coarsening and parallelism with reduction multigrids for hyperbolic Boltzmann transport

S. Dargaville, R. P. Smedley-Stevenson, P. N. Smith, C. C. Pain

AI总结:

针对双曲玻尔兹曼输运方程的约化多重网格并行化难题,提出并行双程CF分裂方法,结合粗网格重分区等优化策略,实现了优于hypre中ℓAIR的求解效率与良好弱缩放性能。

摘要:

约化多重网格近期在双曲问题中展现出良好性能,且无需使用高斯-赛德尔光滑器。将这类方法应用于玻尔兹曼输运方程(BTE)的双曲极限时,在非结构网格上,其计算量随问题规模的增长非常接近$\n\nmathcal{O}(n)$复杂度。这种可扩展性依赖于CF分裂能生成一个易于求逆的$A_\textrm{ff}$块。我们提出了一种并行双程CF分裂方法,旨在得到对角占优的$A_\textrm{ff}$。第一程在对称化的强连接中计算最大独立集;第二程基于$A_\textrm{ff}$的逐行对角占优性,将F点转换为C点。我们发现这种双程CF分裂的性能优于hypre中现有的常用CF分裂方法。\n\n此外,双曲问题中约化多重网格的并行化难度较大,因为我们既需要长程网格转移算子,又需要慢速粗化(二维和三维的粗化速率均约为1/2)。我们发现,设置阶段和求解阶段的良好并行性能取决于多个因素:粗网格的重新分区、粗化过程中减少活跃MPI秩的数量、截断多重网格层级,以及采用GMRES多项式作为粗网格求解器。\n\n我们对比了两种不同约化多重网格的性能:我们此前开发的AIRG,以及hypre中实现的$\n\ne$llAIR。在输运极限下使用AIRG时,我们验证了从2节点到64节点(256核到8196核)的求解弱缩放效率达81%,每核仅对应8.8k个未知量,求解时间最多比hypre中的$\n\ne$llAIR实现短5.9倍。

英文摘要:

Reduction multigrids have recently shown good performance in hyperbolic problems without the need for Gauss-Seidel smoothers. When applied to the hyperbolic limit of the Boltzmann Transport Equation (BTE), these methods result in very close to $\mathcal{O}(n)$ growth in work with problem size on unstructured grids. This scalability relies on the CF splitting producing an $A_\textrm{ff}$ block that is easy to invert. We introduce a parallel two-pass CF splitting designed to give diagonally dominant $A_\textrm{ff}$. The first pass computes a maximal independent set in the symmetrized strong connections. The second pass converts F-points to C-points based on the row-wise diagonal dominance of $A_\textrm{ff}$. We find this two-pass CF splitting outperforms common CF splittings available in hypre. Furthermore, parallelisation of reduction multigrids in hyperbolic problems is difficult as we require both long-range grid-transfer operators and slow coarsenings (with rates of $\sim$1/2 in both 2D and 3D). We find that good parallel performance in the setup and solve is dependent on several factors: repartitioning the coarse grids, reducing the number of active MPI ranks as we coarsen, truncating the multigrid hierarchy and applying a GMRES polynomial as a coarse-grid solver. We compare the performance of two different reduction multigrids, AIRG (that we developed previously) and the hypre implementation of $\ell$AIR. In the streaming limit with AIRG, we demonstrate 81\% weak scaling efficiency in the solve from 2 to 64 nodes (256 to 8196 cores) with only 8.8k unknowns per core, with solve times up to 5.9$\times$ smaller than the $\ell$AIR implementation in hypre.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑