A general class of explicit pseudo--two-step RKN methods on parallel computers (Q1963130)

From MaRDI portal
Revision as of 12:20, 29 May 2024 by ReferenceBot (talk | contribs) (‎Changed an Item)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
A general class of explicit pseudo--two-step RKN methods on parallel computers
scientific article

    Statements

    A general class of explicit pseudo--two-step RKN methods on parallel computers (English)
    0 references
    0 references
    0 references
    0 references
    20 January 2000
    0 references
    The paper proposes a general class of explicit pseudo two-step Runge-Kutta-Nyström methods (EPTRKN methods) requiring only one sequential ordinary differential equation-function evaluation per step for any order of accuracy. The new methods of order from 3 to 10 are shown to be superior, in implementation with fixed stepsize strategy, to the most used sequential and parallel methods. Stability considerations reveal that for any EPTRKN method, the zero-stability requirement is verified. The stability regions of a number of specified methods used in the numerical experiments are effectively large for non-stiff problems. Two strategies are described for the optimal choice of the method parameters and for variable stepsize implementations on parallel computers.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    Runge-Kutta-Nyström methods
    0 references
    parallel computation
    0 references
    non-stiff second-order differential equations
    0 references
    stepsize strategy
    0 references
    stability
    0 references
    stability regions
    0 references
    0 references
    0 references
    0 references
    0 references