Το work with title Preconditioning for solving Hermite collocation by the Bi-CGSTAB by Mathioudakis Emmanouil, Saridakis Ioannis, Papadopoulou Eleni is licensed under Creative Commons Attribution 4.0 International
Bibliographic Citation
Ε.Ν. Mathioudakis, Ε. Ρ. Papadopoulou , Υ. G. Saridakis,.(2006). Preconditioning for solving Hermite Collocation by the Bi-CGSTAB . ACM Transactions on Mathematical Software [online].pp 811-816. Available:http://www.researchgate.net/profile/Yiannis_Saridakis/publication/255575071_Preconditioning_for_solving_Hermite_Collocation_by_the_Bi-CGSTAB/links/544dc44d0cf2bcc9b1d8f2f4.pdf
Explicit pre/post conditioning of the large, sparse and non-symmetric system of equations, arising from the discretization of the Dirichlet Poisson’s Boundary Value Problem (BVP) by the Hermite Collocation method is the problem considered herein. Using the 2-cyclic (red-black) structure of the Collocation coefficient matrix, we investigate the eigenvalue distribution of its preconditioned analogs emerging from its red-black USSOR (UnSymmet- ric SOR) splittings. This analysis, coupled with computational efficiency issues, enables us to justify the choice of Gauss-Seidel (GS) preconditioned schemes as efficient and practical ones, when they used to accelerate the rate of convergence of the Bi-CGSTAB iterative Krylov subspace method. Our results are verified by numerical experiments.