An Improved SSOR-Like Preconditioner for the Non-Hermitian-Positive Definite Linear System with a Dominant Skew-Hermitian Part

Abstract

An improved SSOR-like (ISSOR-like) preconditioner is proposed for the non-Hermitian positive definite linear system with a dominant skew-Hermitian part. The upper and lower bounds on the real and imaginary parts of the eigenvalues of the ISSOR-like preconditioned matrix and the convergence property of the corresponding ISSOR-like iteration method are discussed in depth. Numerical experiments show that the ISSOR-like preconditioner can effectively accelerate preconditioned GMRES.

Keywords

Non-Hermitian positive definiteness / Dominant skew-Hermitian part / Improved SSOR-like (ISSOR-like) / Preconditioner / Eigenvalue distribution / 65F08

References

[1] AxelssonO. A generalized SSOR method. BIT, 1972, 12(4): 443-467

[2] Axelsson, O., Barker, V.A.: Finite Element Solution of Boundary Value Problems: Theory and Computation. Academic Press, Orlando (1984)

[3] BaiZ-Z. A class of modified block SSOR preconditioners for symmetric positive definite systems of linear equations. Adv. Comput. Math., 1999, 10(2): 169-186

[4] BaiZ-Z. Modified block SSOR preconditioners for symmetric positive definite linear systems. Ann. Oper. Res., 2001, 103: 263-282

[5] BaiZ-Z. On SSOR-like preconditioners for non-Hermitian positive definite matrices. Numer. Linear Algebra Appl., 2016, 23(1): 37-60

[6] BaiZ-Z. Respectively scaled HSS iteration methods for solving discretized spatial fractional diffusion equations. Numer. Linear Algebra Appl., 2018, 255e2157

[7] Bai, Z.-Z., Benzi, M., Chen, F.: Modified HSS iteration methods for a class of complex symmetric linear systems. Computing 87(3/4), 93–111 (2010)

[8] BaiZ-Z, GolubGH, LiC-K. Convergence properties of preconditioned Hermitian and skew-Hermitian splitting methods for non-Hermitian positive semidefinite matrices. Math. Comput., 2007, 76(257): 287-298

[9] BaiZ-Z, GolubGH, NgMK. Hermitian and skew-Hermitian splitting methods for non-Hermitian positive definite linear systems. SIAM J. Matrix Anal. Appl., 2003, 24(3): 603-626

[10] BaiZ-Z, GolubGH, NgMK. On successive-overrelaxation acceleration of the Hermitian and skew-Hermitian splitting iterations. Numer. Linear Algebra Appl., 2007, 14(4): 319-335

[11] BaiZ-Z, NgMK. Preconditioners for nonsymmetric block Toeplitz-like-plus-diagonal linear systems. Numer. Math., 2003, 96(2): 197-220

[12] BotchevMA, GolubGH. A class of nonsymmetric preconditioners for saddle point problems. SIAM J. Matrix Anal. Appl., 2006, 27(4): 1125-1149

[13] BotchevMA, KrukierLA. Iterative solution of strongly nonsymmetric systems of linear algebraic equations. Russian Comput. Math. Math. Phys., 1997, 37: 1241-1251

[14] ChenX, TohKC, PhoonKK. A modified SSOR preconditioner for sparse symmetric indefinite linear systems of equations. Int. J. Numer. Meth. Eng., 2006, 65(6): 785-807

[15] DelongMA, OrtegaJM. SOR as a preconditioner. Appl. Numer. Math., 1995, 18(4): 431-440

[16] EisenstatSC. Efficient implementation of a class of preconditioned conjugate gradient methods. SIAM J. Sci. Stat. Comput., 1981, 2(1): 1-4

[17] Hadjidimos, A., Yeyios, A.K.: Some notes on multisplitting methods and $m$-step SSOR preconditioners for linear systems. Linear Algebra Appl. 248, 277–301 (1996)

[18] Harrar, D.L., II., Ortega, J.M.: Optimum $m$-step SSOR preconditioning. Comput. Appl. Math. 24, 195–198 (1988)

[19] KrukierLA, ChikinaLG, BelokonTV. Triangular skew-symmetric iterative solvers for strongly nonsymmetric positive real linear system of equations. Appl. Numer. Math., 2002, 41: 89-105

[20] KrukierLA, KrukierBL, RenZ-R. Generalized skew-Hermitian triangular splitting iteration methods for saddle-point linear systems. Numer. Linear Algebra Appl., 2014, 21: 152-170

[21] KumagaiM, KakitaS, OkamotoY. Performance of Krylov subspace method with SOR preconditioner supported by Eisenstat’s technique for linear system derived from time-periodic FEM. COMPEL, 2019, 38(5): 1641-1654

[22] MengZ-H, LiF-T, XuX-C, HuangD-N, ZhangD-L. Fast inversion of gravity data using the symmetric successive over-relaxation (SSOR) preconditioned conjugate gradient algorithm. Explor. Geophys., 2017, 48(3): 294-304

[23] Peng, X.-F., Xiang, S.-H., Li, W.: The test algorithm and the quasi-optimum factor of SSORPCG. Numer. Math. J. of Chin. Univ. 29(2), 176–185 (2007)

[24] Saad, Y.: Highly parallel preconditioners for general sparse matrices. In: Golub, G.H., Luskin, R.M., Greenbaum, A. (eds.) Recent Advances in Iterative Methods, pp. 165–199. Springer, New York (1994)

[25] Saad, Y.: Iterative Methods for Sparse Linear Systems. Society for Industrial and Applied Mathematics, Philadelphia (2003)

[26] SongS-Z, HuangZ-D. A modified SSOR-like preconditioner for non-Hermitian positive definite matrices. Appl. Numer. Math., 2021, 164: 175-189

[27] StotlandSA, OrtegaJM. Orderings for parallel conjugate gradient preconditioners. SIAM J. Sci. Comput., 1997, 18(3): 854-868

[28] TanX-Y. Shifted SSOR-like preconditioner for non-Hermitian positive definite matrices. Numer. Algorithms, 2017, 75(1): 245-260

[29] WangL, BaiZ-Z. Skew-Hermitian triangular splitting iteration methods for non-Hermitian positive definite linear systems of strong skew-Hermitian parts. BIT, 2004, 44: 363-386

[30] Yamada, S., Ohsaki, I., Ikeuchi, M., Niki, H.: Non-adaptive and adaptive SAOR-CG algorithms. J. Comput. Appl. Math. 12/13, 635–650 (1985)

[31] ZhangJ-L. On SSOR-like preconditioner for saddle point problems with dominant skew-Hermitian part. Int. J. Comput. Math., 2019, 96(4): 782-796

[32] ZhangJ-L. Inexact block SSOR-like preconditioners for non-Hermitian positive definite linear systems of strong Hermitian parts. Appl. Numer. Math., 2021, 165: 598-613

Funding

Natural Science Foundation of Zhejiang Province(LQ24A010016)

National Natural Science Foundation of China(11871430)

RIGHTS & PERMISSIONS

Shanghai University