Parallel ODE-solvers with stepsize control

  • Authors:
  • A. Bellen;R. Vermiglio;M. Zennaro

  • Affiliations:
  • Dipartimento di Scienze Matematiche, Universita`di Trieste, I-34100 Trieste, Italy;Dipartimento di Matematica e Informatica, Universitádi Udine, I-33100 Udine, Italy;Dipartimento di Matematica e Informatica, Universitádi Udine, I-33100 Udine, Italy

  • Venue:
  • Journal of Computational and Applied Mathematics
  • Year:
  • 1990

Quantified Score

Hi-index 7.29

Visualization

Abstract

In this paper we propose a parallel implementation of one-step methods with stepsize control for the numerical solution of IVPs for ODEs of the form y'(t)=f(t, y(t)), y(t"0)=y"0. The proposed implementation is based on the fact that any one-step ODE-method on a mesh {t"0"n), y"0 known. In a previous paper (1989) we introduced a paral iterative algorithm for the approximation of the trajectory (y"0, y"1,..., y"N), in which a block of guessed values (u^0"0 := y"0, u^0"1,..., u^0"N is iterated, concurrently with respect to the index n, until an error proportional to a given iteration tolerance TOL "i"t is reached. Here that parallel algorithm is developed further in order to perform the stepsize control strategy, according to a given step tolerance TOL "s"t. Moreover, an analysis of the optimal ratio between TOL "i"t and TOL "s"t is given. The paper ends with numerical examples and estimations of the attainable speedup.