GPU accelerated spectral finite elements on all-hex meshes

DOI10.1016/J.JCP.2016.08.005zbMATH Open1360.65283arXiv1506.05996OpenAlexW1900162302MaRDI QIDQ525945FDOQ525945

Rajesh Gandham, J.-F. Remacle, T. Warburton

Publication date: 8 May 2017

Published in: Journal of Computational Physics (Search for Journal in Brave)

Abstract: This paper presents a spectral element finite element scheme that efficiently solves elliptic problems on unstructured hexahedral meshes. The discrete equations are solved using a matrix-free preconditioned conjugate gradient algorithm. An additive Schwartz two-scale preconditioner is employed that allows h-independence convergence. An extensible multi-threading programming API is used as a common kernel language that allows runtime selection of different computing devices (GPU and CPU) and different threading interfaces (CUDA, OpenCL and OpenMP). Performance tests demonstrate that problems with over 50 million degrees of freedom can be solved in a few seconds on an off-the-shelf GPU.

Full work available at URL: https://arxiv.org/abs/1506.05996

zbMATH Keywords

GPU computingspectral finite elementshexahedral meshes

Mathematics Subject Classification ID

Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Spectral, collocation and related methods for boundary value problems involving PDEs (65N35)

Cites Work

• Gmsh: A 3-D finite element mesh generator with built-in pre- and post-processing facilities
• Towards a complete FEM-based simulation toolkit on GPUs: unstructured grid finite element geometric multigrid solvers with strong smoothers based on sparse approximate inverses
• High-Order Finite-Differences on Multi-threaded Architectures Using OCCA
• High-Order Methods for Incompressible Fluid Flow
• An overlapping Schwarz method for spectral element solution of the incompressible Navier-Stokes equations
• A stabilized nodally integrated tetrahedral
• Architecting the finite element method pipeline for the GPU
• Assembly of finite element methods on graphics processors
• Finite Element Integration on GPUs
• Finite element assembly strategies on multi‐core and many‐core architectures
• An aggregation-based algebraic multigrid method
• Additive Schwarz methods for the \(p\)-version finite element method
• Hierarchic multigrid iteration strategy for the discontinuous Galerkin solution of the steady Euler equations
• GPU Accelerated Discontinuous Galerkin Methods for Shallow Water Equations
• A GPU accelerated aggregation algebraic multigrid method
• Fully‐automated hex‐dominant mesh generation with directionality control via packing rectangular solid cells
• Title not available (Why is that?)

Cited In (9)

• A Systematic Approach to Constructing Preconditioners for the $hp$-Version Mass Matrix on Unstructured and Hybrid Finite Element Meshes
• Fast Matrix-Free Evaluation of Discontinuous Galerkin Finite Element Operators
• High-performance implementation of discontinuous Galerkin methods with application in fluid flow
• Hybrid multigrid methods for high-order discontinuous Galerkin discretizations
• A GPU-accelerated implicit meshless method for compressible flows
• A Scalable and Robust Vertex-Star Relaxation for High-Order FEM
• Orthogonal bases for vertex-mapped pyramids
• GPU-Accelerated Discontinuous Galerkin Methods on Polytopic Meshes
• GPU-accelerated discontinuous Galerkin methods on hybrid meshes

Uses Software

• Gmsh
• FE-gMG
• OCCA

This page was built for publication: GPU accelerated spectral finite elements on all-hex meshes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q525945)