A fast block low-rank dense solver with applications to finite-element matrices

This article presents a fast solver for the dense "frontal" matrices that arise from the multifrontal sparse elimination process of 3D elliptic PDEs. The solver relies on the fact that these matrices can be efficiently represented as a hierarchically off-diagonal low-rank (HODLR) matrix. To construct the low-rank approximation of the off-diagonal blocks, we propose a new pseudo-skeleton scheme, the boundary distance low-rank approximation, that picks rows and columns based on the location of their corresponding vertices in the sparse matrix graph. We compare this new low-rank approximation method to the adaptive cross approximation (ACA) algorithm and show that it achieves better speedup specially for unstructured meshes. Using the HODLR direct solver as a preconditioner (with a low tolerance) to the GMRES iterative scheme, we can reach machine accuracy much faster than a conventional LU solver. Numerical benchmarks are provided for frontal matrices arising from 3D finite element problems corresponding to a wide range of applications.

[1]  Wolfgang Hackbusch,et al.  Construction and Arithmetics of H-Matrices , 2003, Computing.

[2]  Leslie Greengard,et al.  Fast Direct Methods for Gaussian Processes and the Analysis of NASA Kepler Mission Data , 2014 .

[3]  Santosh S. Vempala,et al.  Adaptive Sampling and Fast Low-Rank Matrix Approximation , 2006, APPROX-RANDOM.

[4]  Sandip,et al.  Gas Turbine Heat Transfer and Cooling Technology, Second Edition , 2012 .

[5]  Shivkumar Chandrasekaran,et al.  A Fast Solver for HSS Representations via Sparse Matrices , 2006, SIAM J. Matrix Anal. Appl..

[6]  Olof B. Widlund,et al.  FETI‐DP, BDDC, and block Cholesky methods , 2006 .

[7]  Eric Darve,et al.  Large-scale stochastic linear inversion using hierarchical matrices , 2013, Computational Geosciences.

[8]  Leslie Greengard,et al.  Fast Direct Methods for Gaussian Processes , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Lexing Ying,et al.  A fast direct solver for elliptic problems on general meshes in 2D , 2012, J. Comput. Phys..

[10]  S. Goreinov,et al.  A Theory of Pseudoskeleton Approximations , 1997 .

[11]  Per-Gunnar Martinsson,et al.  An O(N) algorithm for constructing the solution operator to 2D elliptic boundary value problems in the absence of body loads , 2013, Advances in Computational Mathematics.

[12]  W. Hackbusch,et al.  H 2 -matrix approximation of integral operators by interpolation , 2002 .

[13]  Per-Gunnar Martinsson,et al.  A Fast Randomized Algorithm for Computing a Hierarchically Semiseparable Representation of a Matrix , 2011, SIAM J. Matrix Anal. Appl..

[14]  Per-Gunnar Martinsson,et al.  A Fast Direct Solver for a Class of Elliptic Partial Differential Equations , 2009, J. Sci. Comput..

[15]  W. Hackbusch,et al.  A sparse H -matrix arithmetic: general complexity estimates , 2000 .

[16]  William W. Hager,et al.  Updating the Inverse of a Matrix , 1989, SIAM Rev..

[17]  Jianlin Xia,et al.  Fast algorithms for hierarchically semiseparable matrices , 2010, Numer. Linear Algebra Appl..

[18]  D. Rixen,et al.  FETI‐DP: a dual–primal unified FETI method—part I: A faster alternative to the two‐level FETI method , 2001 .

[19]  Jianlin Xia,et al.  Randomized Sparse Direct Solvers , 2013, SIAM J. Matrix Anal. Appl..

[20]  V. Rokhlin,et al.  A fast direct solver for boundary integral equations in two dimensions , 2003 .

[21]  Eric Darve,et al.  Fast Algorithms for Bayesian Inversion , 2013 .

[22]  Eric Darve,et al.  The black-box fast multipole method , 2009, J. Comput. Phys..

[23]  Bruno Després,et al.  A Domain Decomposition Method for the Helmholtz equation and related Optimal Control Problems , 1996 .

[24]  Kenneth L. Ho,et al.  Hierarchical Interpolative Factorization for Elliptic Operators: Differential Equations , 2016 .

[25]  Y. Saad,et al.  GMRES: a generalized minimal residual algorithm for solving nonsymmetric linear systems , 1986 .

[26]  A. George.  Nested Dissection of a Regular Finite Element Mesh , 1973 .

[27]  Mario Bebendorf,et al.  Hierarchical Matrices: A Means to Efficiently Solve Elliptic Boundary Value Problems , 2008 .

[28]  Lexing Ying.  Fast Algorithms for Boundary Integral Equations , 2009 .

[29]  Frédéric Hecht,et al.  New development in freefem++ , 2012, J. Num. Math..

[30]  Jianlin Xia,et al.  Superfast Multifrontal Method for Large Structured Linear Systems of Equations , 2009, SIAM J. Matrix Anal. Appl..

[31]  Leslie Greengard,et al.  A Fast Direct Solver for Structured Linear Systems by Recursive Skeletonization , 2012, SIAM J. Sci. Comput..

[32]  Timothy A. Davis,et al.  The university of Florida sparse matrix collection , 2011, TOMS.

[33]  Petros Drineas,et al.  CUR matrix decompositions for improved data analysis , 2009, Proceedings of the National Academy of Sciences.

[34]  John K. Reid,et al.  The Multifrontal Solution of Indefinite Sparse Symmetric Linear , 1983, TOMS.

[35]  Shivkumar Chandrasekaran,et al.  A Fast ULV Decomposition Solver for Hierarchically Semiseparable Representations , 2006, SIAM J. Matrix Anal. Appl..

[36]  Eric Darve,et al.  A Kalman filter powered by H2 ‐matrices for quasi‐continuous data assimilation problems , 2014, ArXiv.

[37]  David Gordon Wilson,et al.  The design of high-efficiency turbomachinery and gas turbines , 1984 .

[38]  Patrick R. Amestoy,et al.  Improving Multifrontal methods by means of Low-Rank Approximations techniques , 2012 .

[39]  Joseph W. H. Liu,et al.  The Multifrontal Method for Sparse Matrix Solution: Theory and Practice , 1992, SIAM Rev..

[40]  Jean-Yves L'Excellent,et al.  Improving Multifrontal Methods by Means of Block Low-Rank Representations , 2015, SIAM J. Sci. Comput..

[41]  M. Hestenes,et al.  Methods of conjugate gradients for solving linear systems , 1952 .

[42]  Per-Gunnar Martinsson,et al.  On the Compression of Low Rank Matrices , 2005, SIAM J. Sci. Comput..

[43]  Sergej Rjasanow,et al.  Adaptive Cross Approximation of Dense Matrices , 2000 .

[44]  Steffen Börm,et al.  Data-sparse Approximation by Adaptive ℋ2-Matrices , 2002, Computing.

[45]  Eric Darve,et al.  The Inverse Fast Multipole Method , 2014, ArXiv.

[46]  Alan M. Frieze,et al.  Fast monte-carlo algorithms for finding low-rank approximations , 2004, JACM.

[47]  Ming Gu,et al.  Efficient Algorithms for Computing a Strong Rank-Revealing QR Factorization , 1996, SIAM J. Sci. Comput..

[48]  S. CHANDRASEKARAN,et al.  SOME FAST ALGORITHMS FOR HIERARCHICALLY SEMISEPARABLE MATRICES , 2007 .

[49]  W. Hackbusch.  A Sparse Matrix Arithmetic Based on $\Cal H$-Matrices. Part I: Introduction to ${\Cal H}$-Matrices , 1999, Computing.

[50]  Jean Roman,et al.  SCOTCH: A Software Package for Static Mapping by Dual Recursive Bipartitioning of Process and Architecture Graphs , 1996, HPCN Europe.

[51]  Eric Darve,et al.  An $$\mathcal O (N \log N)$$O(NlogN)  Fast Direct Solver for Partial Hierarchically Semi-Separable Matrices , 2013 .

[52]  Jianlin Xia,et al.  Efficient Structured Multifrontal Factorization for General Large Sparse Matrices , 2013, SIAM J. Sci. Comput..

[53]  W. Hackbusch,et al.  On H2-Matrices , 2000 .

[54]  Boris N. Khoromskij,et al.  A Sparse H-Matrix Arithmetic. Part II: Application to Multi-Dimensional Problems , 2000, Computing.

[55]  Leslie Greengard,et al.  A Fast Direct Solver for High Frequency Scattering from a Large Cavity in Two Dimensions , 2014, SIAM J. Sci. Comput..

[56]  Srinath V. Ekkad,et al.  Gas Turbine Heat Transfer and Cooling Technology , 2012 .

[57]  V. Rokhlin,et al.  A randomized algorithm for the approximation of matrices , 2006 .

[58]  James Bremer,et al.  An adaptive fast direct solver for boundary integral equations in two dimensions , 2009 .

[59]  Bruce M. Irons,et al.  A frontal solution program for finite element analysis , 1970 .

[60]  Per-Gunnar Martinsson,et al.  A direct solver with O(N) complexity for integral equations on one-dimensional domains , 2011, 1105.5372.

[61]  Patrick Amestoy,et al.  A Fully Asynchronous Multifrontal Solver Using Distributed Dynamic Scheduling , 2001, SIAM J. Matrix Anal. Appl..

[62]  Mario Bebendorf,et al.  Approximation of boundary element matrices , 2000, Numerische Mathematik.

[63]  Nathan Halko,et al.  Finding Structure with Randomness: Probabilistic Algorithms for Constructing Approximate Matrix Decompositions , 2009, SIAM Rev..

[64]  V. Rokhlin,et al.  A fast randomized algorithm for the approximation of matrices ✩ , 2007 .