Title :
Stochastic convergence of asynchronous parallel computations for solving systems of linear equations
Author :
Leland, Robert P.
Author_Institution :
Dept. of Electr. Eng., Alabama Univ., Tuscaloosa, AL, USA
Abstract :
We consider the convergence of asynchronous parallel iterative algorithms for systems of linear equations. We assume the processors operate in a Poisson manner. Using Lyapunov functions, we demonstrate a condition for both mean square and almost sure convergence independent of processing rates. Our condition is less conservative than the worst case condition of Chazan and Miranker (1969), and yields results in keeping with asynchronous computation experience
Keywords :
Lyapunov methods; convergence of numerical methods; iterative methods; parallel algorithms; stochastic processes; Lyapunov functions; Poisson manner; asynchronous parallel iterative algorithms; linear equations; sochastic convergence; Computational modeling; Concurrent computing; Convergence; Delay; Iterative algorithms; Lyapunov method; Poisson equations; Stochastic processes; Stochastic systems; Vectors;
Conference_Titel :
Decision and Control, 1993., Proceedings of the 32nd IEEE Conference on
Conference_Location :
San Antonio, TX
Print_ISBN :
0-7803-1298-8
DOI :
10.1109/CDC.1993.325892
