An improved parallel hybrid bi-conjugate gradient method suitable for distributed parallel computing

An improved parallel hybrid bi-conjugate gradient method (IBiCGSTAB(2) method, in brief) for solving large sparse linear systems with nonsymmetric coefficient matrices is proposed for distributed parallel environments. The method reduces five global synchronization points to two by reconstructing the BiCGSTAB(2) method in [G.L.G. Sleijpen, H.A. van der Vorst, Hybrid bi-conjugate gradient methods for CFD problems, in: M. Hafez, K. Oshima (Eds.), Computational Fluid Dynamics Review 1995, John Wiley & Sons Ltd, Chichester, 1995, pp. 457-476] and the communication time required for the inner product can be efficiently overlapped with useful computation. The cost is only slightly increased computation time, which can be ignored, compared with the reduction of communication time. Performance and isoefficiency analysis shows that the IBiCGSTAB(2) method has better parallelism and scalability than the BiCGSTAB(2) method. Numerical experiments show that the scalability can be improved by a factor greater than 2.5 and the improvement in parallel communication performance approaches 60%.

[1]  Laurence T. Yang,et al.  The Improved Quasi-minimal Residual Method on Massively Distributed Memory Computers , 1997, HPCN Europe.

[2]  H. V. D. Vorst,et al.  Reducing the effect of global communication in GMRES( m ) and CG on parallel distributed memory computers , 1995 .

[3]  Eric de Sturler,et al.  A Performance Model for Krylov Subspace Methods on Mesh-Based Parallel Computers , 1996, Parallel Comput..

[4]  L.T. Yang,et al.  The improved BiCGStab method for large and sparse unsymmetric linear systems on parallel distributed memory architectures , 2002, Fifth International Conference on Algorithms and Architectures for Parallel Processing, 2002. Proceedings..

[5]  Laurence T. Yang,et al.  The improved BiCG method for large and sparse linear systems on parallel distributed memory architectures , 2002, Proceedings 16th International Parallel and Distributed Processing Symposium.

[6]  Henk A. van der Vorst,et al.  Bi-CGSTAB: A Fast and Smoothly Converging Variant of Bi-CG for the Solution of Nonsymmetric Linear Systems , 1992, SIAM J. Sci. Comput..

[7]  David Potter.  Computational physics , 1973 .

[8]  R. Freund,et al.  QMR: a quasi-minimal residual method for non-Hermitian linear systems , 1991 .

[9]  Zhiqiang Sheng,et al.  A parallel version of QMRCGSTAB method for large linear systems in distributed parallel environments , 2006, Appl. Math. Comput..

[10]  Zhiqiang Sheng,et al.  An improved bi-conjugate residual algorithm suitable for distributed parallel computing , 2007, Appl. Math. Comput..

[11]  Liu Jie,et al.  An Improved Conjugate Residual Algorithm for Large Symmetric Linear Systems , 2006 .

[12]  Philip E. Gill,et al.  Numerical Linear Algebra and Optimization , 1991 .

[13]  H. Martin Bücker,et al.  A Parallel Version of the Quasi-Minimal Residual Method, Based on Coupled Two-Term Recurrences , 1996, PARA.

[14]  Xuebin Chi,et al.  Multiple search direction conjugate gradient method II: theory and numerical experiments , 2004, Int. J. Comput. Math..

[15]  M. Hafez,et al.  Computational fluid dynamics review 1995 , 1995 .

[16]  Yousef Saad,et al.  Iterative methods for sparse linear systems , 2003 .

[17]  Tongxiang Gu,et al.  Multiple search direction conjugate gradient method I: methods and their propositions , 2004, Int. J. Comput. Math..

[18]  L.T. Yang,et al.  The improved CGS method for large and sparse linear systems on bulk synchronous parallel architectures , 2002, Fifth International Conference on Algorithms and Architectures for Parallel Processing, 2002. Proceedings..