An Eigenwise Parallel Augmented Subspace Method for Eigenvalue Problems

A type of parallel augmented subspace scheme for eigenvalue problems is proposed by using coarse space in the multigrid method. With the help of coarse space in multigrid method, solving the eigenvalue problem in the finest space is decomposed into solving the standard linear boundary value problems and very low dimensional eigenvalue problems. The computational efficiency can be improved since there is no direct eigenvalue solving in the finest space and the multigrid method can act as the solver for the deduced linear boundary value equations. Furthermore, for different eigenvalues, the corresponding boundary value problem and low dimensional eigenvalue problem can be solved in the parallel way since they are independent of each other and there exists no data exchanging. This property means that we do not need to do the orthogonalization in the highest dimensional spaces. This is the main aim of this paper since avoiding orthogonalization can improve the scalability of the proposed numerical method. Some numerical examples are provided to validate the proposed parallel augmented subspace method.

[1]  Andrew Knyazev,et al.  Preconditioned Eigensolvers - an Oxymoron? , 1998 .

[2]  G. Burton.  Sobolev Spaces , 2013 .

[3]  Philippe G. Ciarlet,et al.  The finite element method for elliptic problems , 2002, Classics in applied mathematics.

[4]  Randolph E. Bank,et al.  An optimal order process for solving finite element equations , 1981 .

[5]  Wolfgang Hackbusch,et al.  Multi-grid methods and applications , 1985, Springer series in computational mathematics.

[6]  Hehu Xie,et al.  A multi-level correction scheme for eigenvalue problems , 2011, Math. Comput..

[7]  Hehu Xie,et al.  A full multigrid method for eigenvalue problems , 2016, J. Comput. Phys..

[8]  Andrew V. Knyazev,et al.  A subspace preconditioning algorithm for eigenvector/eigenvalue computation , 1995, Adv. Comput. Math..

[9]  Hehu Xie,et al.  A full multigrid method for eigenvalue problems , 2014, J. Comput. Phys..

[10]  Hehu Xie,et al.  A type of multilevel method for the Steklov eigenvalue problem , 2014 .

[11]  Hamilton-Jacobi Equations,et al.  Multigrid Methods for , 2011 .

[12]  F. Chatelin.  Spectral approximation of linear operators , 2011 .

[13]  E. D'yakonov,et al.  Minimization of the computational labor in determining the first eigenvalues of differential operators , 1980 .

[14]  I. Babuska,et al.  Finite element-galerkin approximation of the eigenvalues and Eigenvectors of selfadjoint problems , 1989 .

[15]  D. Sorensen.  Numerical methods for large eigenvalue problems , 2002, Acta Numerica.

[16]  Wolfgang Hackbusch,et al.  On the Computation of Approximate Eigenvalues and Eigenfunctions of Elliptic Operators by Means of a Multi-Grid Method , 1979 .

[17]  Sophia Blau,et al.  Analysis Of The Finite Element Method , 2016 .

[18]  Walter Greiner,et al.  Quantum Mechanics: An Introduction , 1989 .

[19]  A. Knyazev,et al.  Efficient solution of symmetric eigenvalue problems using multigridpreconditioners in the locally optimal block conjugate gradient method , 2001 .

[20]  D. Sorensen.  IMPLICITLY RESTARTED ARNOLDI/LANCZOS METHODS FOR LARGE SCALE EIGENVALUE CALCULATIONS , 1996 .

[21]  James H. Bramble,et al.  The analysis of multigrid methods , 2000 .

[22]  J. Pasciak,et al.  New convergence estimates for multigrid algorithms , 1987 .

[23]  Jinchao Xu,et al.  Iterative Methods by Space Decomposition and Subspace Correction , 1992, SIAM Rev..

[24]  L. R. Scott,et al.  The Mathematical Theory of Finite Element Methods , 1994 .

[25]  Andrew V. Knyazev,et al.  Toward the Optimal Preconditioned Eigensolver: Locally Optimal Block Preconditioned Conjugate Gradient Method , 2001, SIAM J. Sci. Comput..

[26]  Jinchao Xu.  A new class of iterative methods for nonselfadjoint or indefinite problems , 1992 .