×

The effect of partial pivoting in sparse ordinary differential equation solvers. (English) Zbl 0636.65074

The author considers the effect of partial pivoting in ordinary differential equation solvers that use sparse matrix techniques. Two solvers are considered, one using partial pivoting and LSODES which does not. Results are given for a model problem in fluid flow derived from the one-dimensional Euler equations. These are discretized and solved by the method of lines. The results indicate that partial pivoting is not necessary.
Reviewer: G.J.Cooper

MSC:

65L05 Numerical methods for initial value problems involving ordinary differential equations
65M20 Method of lines for initial value and initial-boundary value problems involving PDEs
76B15 Water waves, gravity waves; dispersion and scattering, nonlinear interaction
35K60 Nonlinear initial, boundary and initial-boundary value problems for linear parabolic equations

Software:

FACSIMILE; DEPAC; MA28; YSMP; LSODE
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Hindmarsh, A. C., Toward a systematized collection of ODE solvers, (Proc. 10th IMACS Congr. on System Simulation and Scientific Computation, Montreal, Quebec, Vol. 1 (Aug. 1982)), 427-432
[2] Eisenstat, S. C., Yale Sparse Matrix Package I. The symmetric codes, (Research Report No. 112 (1977), Dept. of Computer Sciences, Yale Univ: Dept. of Computer Sciences, Yale Univ New Haven, Conn), Yale Sparse Matrix Package II. The unsymmetric codes. Research Report No. 114 · Zbl 0492.65012
[3] Hindmarsh, A. C., LSODE and LSODI. Two new initial value ordinary differential equation solvers, ACM-SIGNUM Newsl., 15, 10-11 (1980)
[4] Thompson, S.; Tuttle, P. G., The solution of several representative stiff problems in an industrial environment: the evolution of an O.D.E. solver, (Proc. Int. Conf. on Stiff Computation, Park City, Utah, Vol. 3 (April 1982))
[5] Curtis, A. R., The FACSIMILE numerical integrator for stiff initial value problems, (Gladwell, I.; Sayers, D. K., Computational Techniques for Ordinary Differential Equations (1980), Academic Press: Academic Press London) · Zbl 0454.65055
[6] R. L. Cox, LS0D28—A variant of LSODE for systems having large sparse Jacobian matrix. Report K/CSD-16; R. L. Cox, LS0D28—A variant of LSODE for systems having large sparse Jacobian matrix. Report K/CSD-16
[7] Duff, I. S., MA28—A set of FORTRAN subroutines for sparse unsymmetric linear equations, (AERE Report R.8730 (1979), AERE: AERE Harwell, Oxon)
[8] Thompson, S., Remarks on the choice of a stiff ordinary differential equation solver, Comput. Math. Applic., 12A, 1125-1141 (1986) · Zbl 0636.65073
[9] Thompson, S.; Tuttle, P. G., Benchmark fluid flow problems for continuous simulation languages, Comput. Math. Applic., 12A, 345-351 (1986) · Zbl 0592.76039
[10] Carver, M. B., Pseudo-characteristic method of lines solution of first order hyperbolic equation systems. Advances in computer methods for partial differential equations III., (Proc. 3rd IMACS Int. Symp. on Computer Methods for Partial Differential Equations. Proc. 3rd IMACS Int. Symp. on Computer Methods for Partial Differential Equations, Bethlehem, Pa (1979)), 227-230
[11] Shampine, L. F.; Watts, H. A., DEPAC—Design of a user oriented package of ODE solvers, (Report SAND 79-2374 (1980), Sandia Labs: Sandia Labs Albuquerque, N.M) · Zbl 0407.68036
[12] Thompson, S., A quantitative assessment of the effect of partial pivoting in sparse ordinary differential equation solvers, (Report ORNL-6207 (1986), Oak Ridge National Lab: Oak Ridge National Lab Oak Ridge, Tenn) · Zbl 0636.65074
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.