GPU performance analysis of a nodal discontinuous Galerkin method for acoustic and elastic models

Abstract

Finite element schemes based on discontinuous Galerkin methods possess features amenable to massively parallel computing accelerated with general purpose graphics processing units (GPUs). However, the computational performance of such schemes strongly depends on their implementation. In the past, several implementation strategies have been proposed. They are based exclusively on specialized compute kernels tuned for each operation, or they can leverage BLAS libraries that provide optimized routines for basic linear algebra operations. In this paper, we present and analyze up-to-date performance results for different implementations, tested in a unified framework on a single NVIDIA GTX980 GPU. We show that specialized kernels written with a one-node-per-thread strategy are competitive for polynomial bases up to the fifth and seventh degrees for acoustic and elastic models, respectively. For higher degrees, a strategy that makes use of the NVIDIA cuBLAS library provides better results, able to reach a net arithmetic throughput 35.7% of the theoretical peak value.

arXiv commentPaper submitted to Computers & Geosciences, 21 pages

References in corpus (8)

• Nodal Discontinuous Galerkin Methods on Graphics Processors
• PyFR: An Open Source Framework for Solving Advection-Diffusion Type Problems on Streaming Architectures using the Flux Reconstruction Approach
• OCCA: A unified approach to multi-threading languages
• GPU-accelerated discontinuous Galerkin methods on hybrid meshes
• GPU Accelerated Discontinuous Galerkin Methods for Shallow Water Equations
• Discontinuous Galerkin methods on graphics processing units for nonlinear hyperbolic conservation laws
• Nodal Discontinuous Galerkin Simulations for Reverse-Time Migration on GPU Clusters
• GPU-accelerated Bernstein-Bezier discontinuous Galerkin methods for wave problems

Cited by in corpus (10)

• Fast matrix-free evaluation of discontinuous Galerkin finite element operators
• Estimation of groundwater storage from seismic data using deep learning
• An entropy stable discontinuous Galerkin method for the shallow water equations on curvilinear meshes with wet/dry fronts accelerated by GPUs
• Weight-adjusted discontinuous Galerkin methods: matrix-valued weights and elastic wave propagation in heterogeneous media
• An implementation of tensor product patch smoothers on GPU
• GPU-accelerated Bernstein-Bezier discontinuous Galerkin methods for wave problems
• A hybridizable discontinuous Galerkin method with characteristic variables for Helmholtz problems
• Weight-adjusted discontinuous Galerkin methods: curvilinear meshes
• Multilevel Interior Penalty Methods on GPUs
• Reduced storage nodal discontinuous Galerkin methods on semi-structured prismatic meshes