Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

A two-level GPU-accelerated incomplete LU preconditioner for general sparse linear systems

Journal Article · · International Journal of High Performance Computing Applications

Here, this paper presents a parallel preconditioning approach based on incomplete LU (ILU) factorizations in the framework of Domain Decomposition (DD) for general sparse linear systems. We focus on distributed memory parallel architectures, specifically, those that are equipped with graphic processing units (GPUs). In addition to block-Jacobi, we present general purpose two-level ILU Schur complement-based approaches, where different strategies are presented to solve the coarse-level reduced system. These strategies are combined with modified ILU methods in the construction of the coarse-level operator, in order to effectively remove smooth errors by targeting an algebraically smooth vector. We leverage available GPU-based sparse matrix kernels to accelerate the setup and the solve phases of the proposed ILU preconditioner. We evaluate the efficiency of the proposed methods as a smoother for algebraic multigrid (AMG) and as a preconditioner for Krylov subspace methods on challenging anisotropic diffusion problems and a collection of general sparse matrices.

Research Organization:
Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA)
Grant/Contract Number:
AC52-07NA27344
OSTI ID:
2537970
Alternate ID(s):
OSTI ID: 2522848
Report Number(s):
LLNL--JRNL-813686; 1021773
Journal Information:
International Journal of High Performance Computing Applications, Journal Name: International Journal of High Performance Computing Applications Journal Issue: 3 Vol. 39; ISSN 1094-3420
Publisher:
SAGECopyright Statement
Country of Publication:
United States
Language:
English

References (65)

Reducing communication in algebraic multigrid using additive variants: REDUCING COMMUNICATION IN AMG WITH ADDITIVE VARIANTS journal February 2014
pARMS: a parallel version of the algebraic recursive multilevel solver journal January 2003
Convergence of a balancing domain decomposition by constraints and energy minimization journal January 2003
FETI-DP: a dual-primal unified FETI method?part I: A faster alternative to the two-level FETI method
  • Farhat, Charbel; Lesoinne, Michel; LeTallec, Patrick
  • International Journal for Numerical Methods in Engineering, Vol. 50, Issue 7 https://doi.org/10.1002/nme.76
journal January 2001
hypre: A Library of High Performance Preconditioners
  • Falgout, Robert D.; Yang, Ulrike Meier; Goos, Gerhard
  • Computational Science — ICCS 2002: International Conference Amsterdam, The Netherlands, April 21–24, 2002 Proceedings, Part III https://doi.org/10.1007/3-540-47789-6_66
book April 2002
A Distributed CPU-GPU Sparse Direct Solver book January 2014
Solving Sparse Linear Systems on NVIDIA Tesla GPUs
  • Wang, Mingliang; Klie, Hector; Parashar, Manish
  • Computational Science – ICCS 2009 -- 9th International Conference Baton Rouge, LA, USA, May 25-27, 2009 Proceedings, Part I https://doi.org/10.1007/978-3-642-01970-8_87
conference May 2009
Partitioning sparse rectangular matrices for parallel processing book January 1998
Abstract robust coarse spaces for systems of PDEs via generalized eigenproblems in the overlaps journal August 2013
Multilevel ILU decomposition journal June 1999
GPU-accelerated preconditioned iterative linear solvers journal October 2012
Experimental study of ILU preconditioners for indefinite matrices journal December 1997
Comparison of multigrid and incomplete LU shifted-Laplace preconditioners for the inhomogeneous Helmholtz equation journal May 2006
Preconditioning Helmholtz linear systems journal April 2010
A two-level ILU preconditioner for electromagnetic applications journal January 2017
A GPU accelerated aggregation algebraic multigrid method journal November 2014
MFEM: A modular finite element methods library journal January 2021
Multigrid reduction preconditioning framework for coupled processes in porous and fractured media journal December 2021
Solving lattice QCD systems of equations using mixed precision solvers on GPUs journal September 2010
Cucheb: A GPU implementation of the filtered Lanczos procedure journal November 2017
Accelerating sparse Cholesky factorization on GPUs journal November 2016
Concurrent number cruncher: a GPU implementation of a general sparse linear solver journal June 2009
An incomplete factorization technique for positive definite linear systems journal May 1980
Parallel multilevel preconditioners journal September 1990
Hypergraph-partitioning-based decomposition for parallel sparse-matrix vector multiplication journal July 1999
A Parallel Direct/Iterative Solver Based on a Schur Complement Approach
  • Gaidamour, J.; Hénon, P.
  • 2008 IEEE 11th International Conference on Computational Science and Engineering (CSE), 2008 11th IEEE International Conference on Computational Science and Engineering https://doi.org/10.1109/CSE.2008.36
conference July 2008
ShyLU: A Hybrid-Hybrid Solver for Multicore Platforms
  • Rajamanickam, Sivasankaran; Boman, Erik G.; Heroux, Michael A.
  • 2012 IEEE International Symposium on Parallel & Distributed Processing (IPDPS), 2012 IEEE 26th International Parallel and Distributed Processing Symposium https://doi.org/10.1109/IPDPS.2012.64
conference May 2012
An Experimental Study of Two-level Schwarz Domain-Decomposition Preconditioners on GPUs conference May 2023
A Hybrid Implementation of Two-Level Domain Decomposition Algorithm for Solving Elliptic Equation on CPU/GPUs conference December 2012
GPU Acceleration of Algebraic Multigrid Preconditioners for Discrete Elliptic Field Problems journal February 2014
A Novel Multigrid Based Preconditioner For Heterogeneous Helmholtz Problems journal January 2006
Spectral Analysis of the Discrete Helmholtz Operator Preconditioned with a Shifted Laplacian journal January 2007
Partitioning Sparse Matrices with Eigenvectors of Graphs journal July 1990
Nested Dissection of a Regular Finite Element Mesh journal April 1973
An Algorithm for Reducing the Bandwidth and Profile of a Sparse Matrix journal April 1976
Compatible Relaxation and Coarsening in Algebraic Multigrid journal January 2010
Iterative Methods for Sparse Linear Systems book January 2003
Multigrid Smoothers for Ultraparallel Computing journal January 2011
Fine-Grained Parallel Incomplete LU Factorization journal January 2015
AmgX: A Library for GPU Accelerated Algebraic Multigrid and Preconditioned Iterative Methods journal January 2015
ViennaCL---Linear Algebra Library for Multi- and Many-Core Architectures journal January 2016
A Rational Function Preconditioner For Indefinite Sparse Linear Systems journal January 2017
A Hierarchical Low Rank Schur Complement Preconditioner for Indefinite Linear Systems journal January 2018
The Eigenvalues Slicing Library (EVSL): Algorithms, Implementation, and Software journal January 2019
Solving the Three-Dimensional High-frequency Helmholtz Equation Using Contour Integration and Polynomial Preconditioning journal January 2020
Combining Machine Learning and Adaptive Coarse Spaces---A Hybrid Approach for Robust FETI-DP Methods in Three Dimensions journal January 2021
On Generalizing the Algebraic Multigrid Framework journal January 2004
Matrix Renumbering ILU: An Effective Algebraic Multilevel ILU Preconditioner for Sparse Matrices journal January 1999
A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs journal January 1998
A Restricted Additive Schwarz Preconditioner for General Sparse Linear Systems journal January 1999
Distributed Schur Complement Techniques for General Sparse Linear Systems journal January 1999
Algebraic Two-Level Preconditioners for the Schur Complement Method journal January 2001
mpibind conference October 2017
Efficient parallel computation of ILU(k) preconditioners conference January 1999
Parallel threshold-based ILU factorization conference January 1997
Modular Finite Element Methods (MFEM) software June 2010
GLVis: OpenGL Finite Element Visualization Tool software June 2010
HiFlow3 – Technical Report on Release 2.0 text January 2017
Acceleration of a parallel BDDC solver by using graphics processing units on subdomains journal November 2022
Incomplete Gaussian Elimination as a Preconditioning for Generalized Conjugate Gradient Acceleration conference November 1983
Constrained Residual Acceleration of Conjugate Residual Methods conference April 2013
Multigrid Reduction for Coupled Flow Problems with Application to Reservoir Simulation conference February 2017
PDSLin User Guide report June 2011
GPU-Accelerated LOBPCG Method with Inexact Null-Space Filtering for Solving Generalized Eigenvalue Problems in Computational Electromagnetics Analysis with Higher-Order FEM journal July 2017
Parallel Implementation of a Two-level Algebraic ILU(k)-based Domain Decomposition Preconditioner journal May 2018

Similar Records

Two-Stage Gauss-Seidel Preconditioners and Smoothers for Krylov Solvers on a GPU Cluster: Preprint
Conference · Tue Feb 08 23:00:00 EST 2022 · OSTI ID:1845268

On performance of Krylov smoothing for fully-coupled AMG preconditioners for VMS resistive MHD
Program Document · Wed Nov 01 00:00:00 EDT 2017 · OSTI ID:1429689

Final Report - Summer Visit 2010
Technical Report · Mon Sep 12 00:00:00 EDT 2011 · OSTI ID:1026473