GPU accelerated spectral finite elements on all-hex meshes

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.

23 pages, 7 figures

References in corpus (1)

• OCCA: A unified approach to multi-threading languages

Cited by in corpus (9)

• Fast matrix-free evaluation of discontinuous Galerkin finite element operators
• GPU-accelerated discontinuous Galerkin methods on hybrid meshes
• Hybrid multigrid methods for high-order discontinuous Galerkin discretizations
• A scalable and robust vertex-star relaxation for high-order FEM
• Finding Hexahedrizations for Small Quadrangulations of the Sphere
• There are 174 Subdivisions of the Hexahedron into Tetrahedra
• Array Program Transformation with Loo.py by Example: High-Order Finite Elements
• Multilevel Interior Penalty Methods on GPUs
• GPU-accelerated discontinuous Galerkin methods on polytopic meshes