A parallel shared-memory implementation of a high-order accurate solution technique for variable coefficient Helmholtz problems

DOI10.1016/J.CAMWA.2019.08.019zbMATH Open1443.65387arXiv1812.07167OpenAlexW2970281300MaRDI QIDQ2004571FDOQ2004571

Authors: Natalie N. Beams, A. Gillman, Russell J. Hewett 

Publication date: 7 October 2020

Published in: Computers & Mathematics with Applications (Search for Journal in Brave)

Abstract: The recently developed Hierarchical Poincar'e-Steklov (HPS) method is a high-order discretization technique that comes with a direct solver. Results from previous papers demonstrate the method's ability to solve Helmholtz problems to high accuracy without the so-called pollution effect. While the asymptotic scaling of the direct solver's computational cost is the same as the nested dissection method, serial implementations of the solution technique are not practical for large scale numerical simulations. This manuscript presents the first parallel implementation of the HPS method. Specifically, we introduce an approach for a shared memory implementation of the solution technique utilizing parallel linear algebra. This approach is the foundation for future large scale simulations on supercomputers and clusters with large memory nodes. Performance results on a desktop computer (resembling a large memory node) are presented.

Full work available at URL: https://arxiv.org/abs/1812.07167

Recommendations

• Parallel solution of the multidimensional Helmholtz/Schrödinger equation using high order methods
• scientific article; zbMATH DE number 1329613
• Parallel controllability methods for the Helmholtz equation
• Parallel Hierarchical Solvers and Preconditioners for Boundary Element Methods
• A parallel multigrid-based preconditioner for the 3D heterogeneous high-frequency Helmholtz equation

zbMATH Keywords

direct solverhigh-order discretizationOpenMPshared-memory parallelizationnumerical partial differential equationsMKL

Mathematics Subject Classification ID

Parallel numerical computation (65Y05) Spectral, collocation and related methods for boundary value problems involving PDEs (65N35)

Cites Work

• SuperLU_DIST
• A fully asynchronous multifrontal solver using distributed dynamic scheduling
• Title not available (Why is that?)
• A sparse matrix arithmetic based on \({\mathfrak H}\)-matrices. I: Introduction to \({\mathfrak H}\)-matrices
• An \(\mathcal O(N\log N)\) fast direct solver for partial hierarchically semi-separable matrices. With application to radial basis function interpolation
• Nested Dissection of a Regular Finite Element Mesh
• Spectral Methods in MATLAB
• Construction and arithmetics of \(\mathcal H\)-matrices
• A direct solver for variable coefficient elliptic PDEs discretized via a composite spectral collocation method
• Superfast Multifrontal Method for Large Structured Linear Systems of Equations
• A divide-and-conquer algorithm for the eigendecomposition of symmetric block-diagonal plus semiseparable matrices
• Algorithms to solve hierarchically semi-separable systems
• A fast direct solver for elliptic problems on general meshes in 2D
• A fast direct solver for a class of elliptic partial differential equations
• A direct solver with \(O(N)\) complexity for integral equations on one-dimensional domains
• An Asynchronous Parallel Supernodal Algorithm for Sparse Gaussian Elimination
• An Adaptive High Order Direct Solution Technique for Elliptic Boundary Value Problems
• A Distributed-Memory Package for Dense Hierarchically Semi-Separable Matrix Computations Using Randomization
• A Direct Solver with $O(N)$ Complexity for Variable Coefficient Elliptic PDEs Discretized via a High-Order Composite Spectral Collocation Method
• A spectrally accurate direct solution technique for frequency-domain scattering problems with variable media
• An accelerated Poisson solver based on multidomain spectral discretization
• A shared- and distributed-memory parallel general sparse direct solver

Cited In (6)

• Parallel solution of the multidimensional Helmholtz/Schrödinger equation using high order methods
• Schwarz methods by domain truncation
• Asymmetric transport computations in Dirac models of topological insulators
• SlabLU: a two-level sparse direct solver for elliptic PDEs
• An Iterative Solver for the HPS Discretization Applied to Three Dimensional Helmholtz Problems
• Sharp bounds on Helmholtz impedance-to-impedance maps and application to overlapping domain decomposition

Uses Software

• SuperLU-DIST
• Matlab
• SuperLU
• MUMPS
• STRUMPACK
• GitHub

This page was built for publication: A parallel shared-memory implementation of a high-order accurate solution technique for variable coefficient Helmholtz problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2004571)