Computing a block incomplete LU preconditioner as the by-product of block left-looking A-biconjugation process

Subject Areas : Statistics

Amin Rafiei ¹ , Behnaz Tolue Haghighi ²

1 - Hakim Sabzevari University
2 - Hakim Sabzevari University

Received: 2017-07-04 Accepted : 2019-01-27 Published : 2019-05-01

Keywords: نسخه پیمایش چپ پیش شرط ساز فاکتورسازی ناکامل قوی, پپیش شرط ساز LU ناقص, نسخه بلوکی فرایند حذفی گاوس,

Abstract :

In this paper, we present a block version of incomplete LU preconditioner which is computed as the by-product of block A-biconjugation process. The pivot entries of this block preconditioner are one by one or two by two blocks. The L and U factors of this block preconditioner are computed separately. The block pivot selection of this preconditioner is inherited from one of the block versions of Gaussian elimination process. The main basis to propose such a block preconditioner is a connection between the Gaussian elimination process and the A-biconjugation algorithm. In the numerical experiment section, we have generated artificial linear systems. Then, we have computed the block and plain versions of this incomplete LU preconditioner. We have used these two preconditioners as the right preconditioner for linear systems. After that, the GMRES(50) method has been applied to solve the right preconditioned systems. The results indicate that the block preconditioner gives fewer number of iterations of GMRES(50) method than the plain version. Therefore, the block preconditioer has a better quality.

References:

[1] Y. Saad, Iterative methods for sparse linear systems. SIAM Publications, Philadelphia, second edition (2003).

 

[2] M. Benzi, and M. Tuma, A sparse approximate inverse preconditioner for nonsymmetric linear systems, SIAM J. Sci. Comput., 19(3) (1998) 968-994.

 

[3] M. Benzi, and M. Tuma, M. A Robust Incomplete Factorization Preconditioner for Positive Definite Matrices, Numer. Linear Alg. Appl., 10 (2003)  385-400.

 

[4] A. Rafiei, B. Tolue, and M. Bollhoefer, Complete pivoting strategy for the left-looking Robust Incomplete Factorization preconditioner, Comput. Math. Appl., 67 (2014)  2055-2070.

 

[5] A. Rafiei, A complete pivoting strategy for the right-looking Robust Incomplete Factorization preconditioner. Comput. Math. Appl., 64 (2012)  2682-2694.

 

[6] Ch. Kruschel, Losen von positiv definiten, unsymmetrischen Matrizen mit Matching-Methoden am Beispiel von Konvektion-Diffusionsgleichungen. Bachelor of Sciences thesis, Technische Universitat Braunschweig (2009).

 

[7] A. Rafiei, M. Bollhoefer  and F. Benkhaldoun, A block version of left-looking AINV preconditioner with one by one or two by two block pivots, Revised for Appl. Math. Comput., (2018).

 

[8] A. Rafiei, Left-looking version of AINV preconditioner with complete pivoting strategy, Linear Alg. Appl., 445 (2014) 103-126.

 

[9] T. Davis, The SuiteSparse Matrix Collection, http://www.cise.ufl.edu/research/sparse/matrices/. Accessed 2017.

 

[10] I. S. Duff, and J. Koster, The design and use of algorithms for permuting large entries to the diagonal of sparse matrices. SIAM J. Matrix Anal. Appl., 20(4) (1999) 889-901.

 

[11] I. S. Duff, and S.  Parlett Strategies for scaling and pivoting for sparse symmetric indefinite problems. SIAM J. Matrix Anal. Appl., 27(2) (2005)  313-340.

 

[12] G. Karypis, and V. Kumar, METIS a Software Package for partitioning Unstructured Graphs and Computing Fill-Reduced Orderings of Sparse Matrices,

http://glaros.dtc.umn.edu/gkhome/metis/metis/download

 

[13] M. Bollhoefer, ILUPACK software package. http://www.icm.tu-bs.de/~bolle/ilupack.

 

[14] The HSL Mathematical Software Library, http://www.hsl.rl.ac.uk.

 

[15] Y. Saad, Sparskit and sparse examples, http://www-users.cs.umn.edu/~saad/software. Accessed 2017.

 

[16] Y. Saad, ITSOL software package. http://www-users.cs.umn.edu/~saad/software.