Abstract
In this article, a new parallel multilevel algebraic recursive generic approximate inverse solver (PMARGAIS) is proposed. PMARGAIS utilizes the parallel modified generic factored approximate sparse inverse (PMGenFAspI) matrix technique designed for shared memory parallel systems. PMARGAIS requires a block independent set reordering scheme, to create a hierarchy of levels. A modified block breadth first search (MBBFS) is proposed for reducing memory requirements and retaining load balancing. The SVD method is used to compute the inverse of the independent blocks that are formed from the reordering scheme, and computes accurately the Schur complement that is used as a coefficient matrix on the next level, resulting in a hybrid direct-iterative method for large linear systems. The solution of the linear system at the last level is performed with the parallel explicit preconditioned BiCGSTAB method in conjunction with the PMGenFAspI matrix. The parallelization of the proposed methods uses the vector units of modern CPUs. Implementation details are provided and numerical results are given demonstrating the applicability and effectiveness of the proposed schemes.
Similar content being viewed by others
References
Arabnia HR, Thapliyal H, Vinod AP (2006) Combined integer and floating point multiplication architecture (CIFM) for FPGAs and its reversible logic implementation. In: 49th IEEE International Midwest Symposium on Circuits and Systems (MWSCAS’06), San Juan, Puerto Rico, August 6–9, pp 148–154
Axelsson O (1996) Iterative solution methods. Cambridge University Press, Cambridge
Benzi M, Meyer CD, Tuma M (1996) A sparse approximate inverse preconditioner for the conjugate gradient method. SIAM J Sci Comput 17(5):1135–1149
Botta EFF, van der Ploeg A, Wubs FW (1996) Nested grids ILU- decomposition (NGILU). J Comp Appl Math 66:515–526
Botta EFF, Wubs W (1997) MRILU: it’s the preconditioning that counts. Technical Report W-9703, Department of Mathematics, University of Groningen, The Netherlands
Chapman B, Jost G, Van Der Pas R (2008) Using OpenMP: portable shared memory parallel programming. The MIT Press, Cambridge
Chow E (2001) Parallel implementation and practical use of sparse approximate inverses with a priori sparsity patterns. Int J High Perf Comput Appl 15:56–74
Chow E (2000) A priori sparsity patterns for parallel sparse approximate inverse preconditioners. SIAM J Sci Comput 21:1804–1822
Davis TA, Hu Y (2011) The University of Florida sparse matrix collection. ACM Trans Math Softw (TOMS) 38(1):1–25
Filelis-Papadopoulos CK, Gravvanis GA (2016) A class of generic factored and multilevel recursive approximate inverse techniques for solving general sparse systems. Eng Comp 33(1):74–99
Golub GH, Reinsch C (1970) Singular value decomposition and least squares solutions. In: Wilkinson JH, Reinsch C (eds) Handbook for automatic computation, vol. 2 (Linear Algebra). Springer-Verlag, New York, pp 134–151
Golub GH, Van Loan CF (1996) Matrix computations, 3rd edn. Johns Hopkins University Press, Baltimore
Gravvanis GA (2009) High performance inverse preconditioning. Arch Comput Meth Engin 16(1):77–108
Gravvanis GA, Filelis-Papadopoulos CK, Matskanidis PI (2014) Algebraic multigrid methods based on generic approximate banded inverse matrix techniques. Comput Model Eng Sci (CMES) 100(4):323–345
Grote MJ, Huckle T (1997) Parallel preconditioning with sparse approximate inverses. SIAM J Sci Comput 18(3):838–853
Grote MJ, Huckle T (1995) Effective parallel preconditioning with sparse approximate inverses. In: Proceedings of SIAM Conference on Parallel Processing for Scientific Computing, SIAM, pp 466–471
Intel Volume 1. Basic Architecture: http://www.c-jump.com/CIS77/reference/Intel/CIS77_24319002/index.html
Kolotolina YuL, Yeremin YuA (1993) Factorized sparse approximate inverse preconditionings. I. Theory. SIAM J Matrix Anal Appl 14:45–58
Manguoglu M (2011) A domain-decomposing parallel sparse linear system solver. J Comput Appl Math 236(3):319–325
Meijerink JA, Van der Vorst HA (1977) An iterative method for linear systems of which the coefficient is a symmetric M-matrix. Math Comput 31:148–162
Ruge A, Stuben K (1987) Algebraic multigrid. In: McCormick (ed) Multigrid methods. Front Appl Math 3(4) SIAM
Saad Y (1994) ILUT: a dual threshold incomplete LU factorization. Num Linear Algebra Appl 1(4):387–402
Saad Y, Suchomel B (2002) ARMS: an algebraic recursive multilevel solver for general sparse linear systems. Num Linear Algebra Appl 9(5):359–378
Saad Y, Zhang J (1999) BILUTM: a domain-based multilevel block ILUT preconditioner for general sparse matrices. SIAM J Matrix Anal Appl 21:279–299
Thapliyal H, Arabnia HR (2006) Reversible programmable logic array (RPLA) using Fredkin and Feynman gates for industrial electronics and applications. In: Proceedings of the International Conference on Computer Design and Conference on Computing in Nanotechnology (CDES’06), Las Vegas, USA, June 26–29, ISBN #: 1-60132-009-4. http://arxiv.org/abs/cs/0609029, pp 70–74
Thapliyal H, Srinivas MB, Arabnia HR (2005) Reversible logic synthesis of half, full and parallel subtractors. In: Proceedings of the International Conference on Embedded Systems and Applications, ESA’05, June, Las Vegas, pp 165–181
Trottenberg U, Osterlee CW, Schuller A (2000) Multigrid. Academic Press, Cambridge
Van der Vorst HA (1992) Bi-CGSTAB: a fast and smoothly converging variant of Bi-CG for the solution of nonsymmetric linear systems. SIAM J Sci Stat Comput 13(2):631–644
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
The PEPBiCGSTAB method, using AVX units, is described by the following algorithmic scheme:
where fmadd(xr3, xr1, xr2) is the fused multiply add operation \({xr3}={xr3}+{xr1}*{xr2}\), where xr1, xr2 and xr3 are vectors consisting of four double-precision floating point numbers.
Rights and permissions
About this article
Cite this article
Makaratzis, A.T., Filelis-Papadopoulos, C.K. & Gravvanis, G.A. Parallel multilevel recursive approximate inverse techniques for solving general sparse linear systems. J Supercomput 72, 2259–2282 (2016). https://doi.org/10.1007/s11227-016-1728-5
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-016-1728-5