GPU algorithms for Efficient Exascale Discretizations

Center for Efficient Exascale Discretizations, Exascale Computing Project, DOE, ceed.exascaleproject.org. Kolev, 2021, Efficient exascale discretizations: High-order finite element methods, Int. J. HPC App., 1 Kreiss, 1972, Comparison of accurate methods for the integration of hyperbolic problems, Tellus, 24, 199, 10.3402/tellusa.v24i3.10634 Babuška, 1994, The p and h−p versions of the finite element method, basic principles and properties, SIAM Rev., 36, 578, 10.1137/1036141 Orszag, 1980, Spectral methods for problems in complex geometry, J. Comput. Phys., 37, 70, 10.1016/0021-9991(80)90005-4 Gottlieb, 1977 Arndt, 2020, Exadg: High-order discontinuous Galerkin for the exa-scale, 189 Bello-Maldonado, 2019, Scalable low-order finite element preconditioners for high-order spectral element Poisson solvers, SIAM J. Sci. Comput., 41, S2, 10.1137/18M1194997 Canuto, 2010, Finite-element preconditioning of g-NI spectral methods, SIAM J. Sci. Comput., 31, 4422, 10.1137/090746367 Moxey, 2020, Efficient matrix-free high-order finite element evaluation for simplicial elements, SIAM J. Sci. Comput., 42, C97, 10.1137/19M1246523 Sun, 2020, A study of vectorization for matrix-free finite element methods, Int. J. High Perform. Comput. Appl., 34, 629, 10.1177/1094342020945005 Anderson, 2020, MFEM: A modular finite element library, Comput. Math. Appl. Kronbichler, 2018, A performance comparison of continuous and discontinuous Galerkin methods with fast multigrid solvers, SIAM J. Sci. Comput., 40, A3423, 10.1137/16M110455X Kronbichler, 2019, Multigrid for matrix-free high-order finite element computations on graphics processors, ACM Trans. Parallel Comput., 6, 1, 10.1145/3322813 Lottes, 2005, Hybrid multigrid/Schwarz algorithms for the spectral element method, J. Sci. Comput., 24, 45, 10.1007/s10915-004-4787-3 Fischer, 2020, Scalability of high-performance PDE solvers, Int. J. HPC App., 34, 562 Brown, 2021, libCEED: Fast algebra for high-order element-based discretizations, J. Open Source Softw., 6, 2945, 10.21105/joss.02945 Abdelfattah, 2021 Karniadakis, 2005 Vos, 2010, From h to p efficiently: Implementing finite and spectral/hp element methods to achieve optimal performance for low-and high-order discretisations, J. Comput. Phys., 229, 5161, 10.1016/j.jcp.2010.03.031 Ainsworth, 2011, Bernstein-Bézier Finite elements of arbitrary order and optimal assembly procedures, SIAM J. Sci. Comput., 33, 3087, 10.1137/11082539X Kirby, 2011, Fast simplicial finite element algorithms using Bernstein polynomials, Numer. Math., 117, 631, 10.1007/s00211-010-0327-2 Swirydowicz, 2019, Acceleration of tensor-product operations for high-order finite element methods, Int. J. High Perform. Comput. Appl., 33, 735, 10.1177/1094342018816368 Medina, 2014 MAGMA: Matrix Algebra on GPU and Multicore Architectures, icl.utk.edu/magma. Abdelfattah, 2016, High-performance tensor contractions for GPUs, 108 N. Beams, A. Abdelfattah, S. Tomov, J. Dongarra, T. Kolev, Y. Dudouit, High-Order Finite Element Method using Standard and Device-Level Batch GEMM on GPUs, in: 11th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems,Proceedings. To Appear, 2020. Hornung, 2014 Chalmers, 2020 Medina, 2015 Chalmers, 2020 N. Chalmers, T. Warburton, streamParanumal: Streaming Microbenchmarks for High-order Finite Element Methods, URL github.com/paranumal/streamparanumal. 2020 Fischer, 2021 Melander, 2020 2020 2020 P.F. Fischer, K. Heisey, M. Min, Scaling limits for PDE-based simulation, in: 22nd AIAA Computational Fluid Dynamics Conference, 2015, p. 3049. Deville, 2002 Otten, 2016, An MPI/OpenACC implementation of a high order electromagnetics solver with GPUDirect communication, Int. J. High Perform. Comput. Appl., 30, 320, 10.1177/1094342015626584 Gong, 2016, Nekbone performance on GPUs with OpenACC and CUDA fortran implementations, special issue on sustainability on ultrascale computing systems and applications, J. Supercomput., 72, 4160, 10.1007/s11227-016-1744-5 Otero, 2019, OpenACC Acceleration for the PN−PN−2 algorithm in Nek5000, J. Parallel Distrib. Comput., 132, 69, 10.1016/j.jpdc.2019.05.010 Fischer, 1998, Projection techniques for iterative solution of Ax̲=b̲ with successive right-hand sides, Comput. Methods Appl. Mech. Engrg., 163, 193, 10.1016/S0045-7825(98)00012-7 Austin, 2020 2020 Y.-H. Lan, P. Fischer, E. Merzari, M. Min, All-hex meshing strategies for densely packed spheres, in: The 29th International Meshing Roundtable, 2021. Anderson, 2018, High-order multi-material ALE hydrodynamics, SIAM J. Sci. Comput., 40, B32, 10.1137/17M1116453 Beckingsale, 2019, Umpire: Application-focused management and coordination of complex hierarchical memory, IBM J. Res. Dev., 1 Dobrev, 2012, High-order curvilinear finite element methods for Lagrangian hydrodynamics, SIAM J. Sci. Comput., 34, B606, 10.1137/120864672 Dobrev, 2016, Multi-material closure model for high-order finite element Lagrangian hydrodynamics, Internat. J. Numer. Methods Engrg., 82, 689, 10.1002/fld.4236 2020 Bello-Maldonado, 2020, A matrix-free hyperviscosity formulation for high-order ALE hydrodynamics, Comput. Fluids, 10.1016/j.compfluid.2020.104577 Dobrev, 2019, The target-matrix optimization paradigm for high-order meshes, SIAM J. Sci. Comput., 41, B50, 10.1137/18M1167206 Dobrev, 2020, Simulation-driven optimization of high-order meshes in ALE hydrodynamics, Comput. Fluids, 208, 10.1016/j.compfluid.2020.104602 Anderson, 2015, Monotonicity in high-order curvilinear finite element arbitrary Lagrangian–Eulerian remap, Internat. J. Numer. Methods Engrg., 77, 249, 10.1002/fld.3965 Anderson, 2017, High-order local maximum principle preserving (MPP) discontinuous Galerkin finite element method for the transport equation, J. Comput. Phys., 334, 102, 10.1016/j.jcp.2016.12.031 Hajduk, 2020, Matrix-free subcell residual distribution for Bernstein finite element discretizations of linear advection equations, Comput. Methods Appl. Mech. Engrg., 359, 10.1016/j.cma.2019.112658 Carson, 2019 Barton, 2018 Gupta, 1972, A method of computing numerically integrated stiffness matrices, Internat. J. Numer. Methods Engrg., 5, 83, 10.1002/nme.1620050108 Gupta, 1983, Efficient numerical integration of element stiffness matrices, Internat. J. Numer. Methods Engrg., 19, 1410, 10.1002/nme.1620190910