Extending DD-αAMG on heterogeneous machines
Gustavo Ramirez-Hidalgo, Lianhua He and Ke-Long Zhang
多网格求解器是现代科学计算模拟的标准。 领域分解基于聚合的代数多网格,也称为DD-αAMG求解器,是晶格量子色动力学的代数多网格求解器的成功实现。 它的CPU实现使得在某些特定的离散化中构建模拟成为可能,否则计算不可行,并且它还推动了该地区其他代数多网格求解器的开发和改进。 从已经部分通过CUDA移植的DD-αAMG的现有版本到在Nvidia GPU上运行多网格求解器的一些最高级别的操作,我们通过使用HIP在ORISE超级计算机上运行来翻译CUDA代码。 此外,我们还扩展了DD-αAMG中可用的平滑器,特别注意Richardson平滑,在我们的数值实验中,多网格求解器比使用GCR平滑更快,只有10个
Multigrid solvers are the standard in modern scientific computing simulations. Domain Decomposition Aggregation-Based Algebraic Multigrid, also known as the DD-αAMG solver, is a successful realization of an algebraic multigrid solver for lattice quantum chromodynamics. Its CPU implementation has made it possible to construct, for some particular discretizations, simulations otherwise computationally unfeasible, and furthermore it has motivated the development and improvement of other algebraic m...