×

Effective numerical algorithms for the solution of algebraic systems arising in spectral methods. (English) Zbl 0754.65031

The authors first present the weak formulation of the spectral collocation method for elliptic boundary value problems on a rectangular domain with Dirichlet and Neumann conditions. If Legendre nodes and weights are used it is shown that the weak formulation naturally leads to symmetric matrices, whereas the more popular approach by the pointwise strong formulation yields unsymmetric spectral matrices.
Next the solution of the algebraic systems with symmetric and unsymmetric spectral matrices is considered and especially the aspects of preconditioning and vectorization are discussed. The effects and results are outlined for several examples, and comparisons are made between the methods of Richardson, CGS, BI-CGSTAB and GMRES obtained on four different supercomputers.
Finally, the multidomain approach is considered in which the given domain is splitted into subregions, where the spectral method is combined with an iteration with respect to interface conditions.

MSC:

65F10 Iterative numerical methods for linear systems
65N35 Spectral, collocation and related methods for boundary value problems involving PDEs
65N55 Multigrid methods; domain decomposition for boundary value problems involving PDEs
35J25 Boundary value problems for second-order elliptic equations
65Y10 Numerical algorithms for specific classes of architectures

Software:

CGS
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Canuto, C.; Giberti, C., Parallelism in a highly accurate algorithm for turbulence simulation (1988), IAN, CNR: IAN, CNR Pavia
[2] Canuto, C.; Hussaini, M. Y.; Quarteroni, A.; Zang, T. A., Spectral Methods in Fluid Dynamics (1988), Springer: Springer New York · Zbl 0658.76001
[3] Canuto, C.; Pietra, P., Boundary and interface conditions within finite element preconditioners for spectral method, (Rept. No 636 (1989), IAN, Pavia University: IAN, Pavia University Pavia), also: J. Comput. Phys. (to appear)
[4] Deville, M.; Mund, E., Chebishev pseudospectral solution of second order elliptic equations with finite element preconditioning, J. Comput. Phys., 60, 517-533 (1985) · Zbl 0585.65073
[5] Deville, M.; Mund, E., Finite element preconditioning for pseudospectral solutions of elliptic problems, SIAM J. Statist. Comput., 2, 311-342 (1990) · Zbl 0701.65075
[6] Dongarra, J. J.; Gustavson, F. G.; Karp, A., Implementing linear algebra algorithms for dense matrices on a vector pipeline machine, SIAM Rev., 26, 91-112 (1984) · Zbl 0539.65009
[7] Funaro, D., A variational formulation for the Chebyshev pseudospectral approximation of Neumann problems, SIAM J. Numer. Anal., 27, 695-703 (1990) · Zbl 0702.65090
[8] Golub, G.; van Loan, C. F., Matrix Computation (1989), The Johns Hopkins University Press: The Johns Hopkins University Press Baltimore, MD
[9] Gottlieb, D.; Orszag, S. A., Numerical Analysis of Spectral Methods: Theory and Applications (1977), SIAM-CBMS: SIAM-CBMS Philadelphia, PA · Zbl 0412.65058
[10] Hockney, R. W.; Jesshope, C. R., Parallel Computers, 2 (1988), Adam Hilger: Adam Hilger Philadelphia, PA · Zbl 0641.68001
[11] Morchoisne, Y., Résolution deséquations de Navier-Stokes par une méthode pseudospectrale en espace-temps, Rech. Aerosp., 5, 293-306 (1979) · Zbl 0418.76026
[12] Orszag, S. A., Spectral methods for problems in complex geometries, J. Comput. Phys., 37, 70-92 (1980) · Zbl 0476.65078
[13] Quarteroni, A.; Sacchi Landriani, G., Domain decomposition preconditioners for the spectral collocation methods, J. Sci. Comput., 3, 45-75 (1988) · Zbl 0675.65116
[14] Quarteroni, A.; Zampieri, E., Finite element preconditioning of spectral approximations to elliptic problems (1991), Minnesota Supercomputer Institute Report: Minnesota Supercomputer Institute Report Minneapolis, MN
[15] Saad, Y.; Schultz, M. H., GMRES: a generalized minimal residual algorithm for solving nonsymmetric linear systems, SIAM J. Sci. Statist. Comput., 7, 856-869 (1986) · Zbl 0599.65018
[16] Saleri, F., Domain decomposition for finite elements and their implementation on a vector multiprocessor, IMPACT, 2, 99-124 (1990)
[17] Sonneveld, P., CGS: a fast Lanczos-type solver for nonsymmetric linear systems, SIAM J. Sci. Statist. Comput., 10, 36-52 (1989) · Zbl 0666.65029
[18] van der Vorst, H. A., Bi-CGSTAB: a fast and smoothly converging variant of Bi-CG for the solutionof nonsymmetric linear systems, ((1990), Utrecht University, Department of Mathematics: Utrecht University, Department of Mathematics Utrecht), also: SIAM J. Sci. Statist. Comput. (to appear) · Zbl 0761.65023
[19] Young, D. M., Iterative Solution of Large Linear System (1971), Academic Press: Academic Press New York · Zbl 0204.48102
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.