A Parallel Jacobi-Davidson-type Method for Solving Large Generalized Eigenvalue Problems in Magnetohydrodynamics

We study the solution of generalized eigenproblems generated by a model which is used for stability investigation of tokamak plasmas. The eigenvalue problems are of the form $A x = \lambda B x$, in which the complex matrices A and B are block-tridiagonal, and B is Hermitian positive definite. The Jacobi--Davidson method appears to be an excellent method for parallel computation of a few selected eigenvalues because the basic ingredients are matrix vector products, vector updates, and inner products. The method is based on solving projected eigenproblems of order typically less than 30. We apply a complete block LU decomposition in which reordering strategies based on a combination of block cyclic reduction and domain decomposition result in a well-parallelizable algorithm. One decomposition can be used for the calculation of several eigenvalues. Spectral transformations are presented to compute certain interior eigenvalues and their associated eigenvectors. The convergence behavior of several variants of the Jacobi--Davidson algorithm is examined. Special attention is paid to the parallel performance, memory requirements, and prediction of the speed-up. Numerical results obtained on a distributed memory Cray T3E are shown.

[1]  Gerard L. G. Sleijpen,et al.  Jacobi-Davidson Style QR and QZ Algorithms for the Reduction of Matrix Pencils , 1998, SIAM J. Sci. Comput..

[2]  Stefaan Poedts,et al.  Computing the damping and destabilization of global Alfvén waves in tokamaks , 1991 .

[3]  Margreet Nool,et al.  Calculation of resistive magnetohydrodynamic spectra in tokamaks , 1999 .

[4]  Gerard L. G. Sleijpen,et al.  A Jacobi-Davidson Iteration Method for Linear Eigenvalue Problems , 1996, SIAM J. Matrix Anal. Appl..

[5]  Margreet Nool,et al.  A parallel Jacobi-Davidson method for solving generalized eigenvalue problems in linear magnetohydrodynamics , 1997 .

[6]  Gene H. Golub,et al.  Matrix computations , 1983 .

[7]  D. R. Fokkema,et al.  Jacobi-Davidson methods for generalized MHD-eigenvalue problems , 1995 .

[8]  Gene H. Golub,et al.  Matrix computations (3rd ed.) , 1996 .

[9]  Henk A. van der Vorst,et al.  Cracking large-scale eigenvalue problems, part I: algorithms , 1996 .

[10]  Richard Barrett,et al.  Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods , 1994, Other Titles in Applied Mathematics.

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

[12]  J. P. Goedbloed,et al.  CASTOR: Normal-Mode Analysis of Resistive MHD Plasmas☆ , 1998 .

[13]  van der,et al.  Reordering strategies and LU-decomposition of block tridiagonal matrices for parallel processing , 1996 .

[14]  James Demmel,et al.  LAPACK Users' Guide, Third Edition , 1999, Software, Environments and Tools.

[15]  W. Kerner,et al.  Damping of global alfven waves in tokamaks due to resonant absorption , 1992 .

[16]  Auke van der Ploeg,et al.  Parallel Jacobi-Davidson for Solving Generalized Eigenvalue Problems , 1998, VECPAR.

[17]  Chao Yang,et al.  ARPACK users' guide - solution of large-scale eigenvalue problems with implicitly restarted Arnoldi methods , 1998, Software, environments, tools.

[18]  Henk A. van der Vorst,et al.  Cracking large-scale eigenvalue problems, part II: implementations , 1996 .