Parallel implementations of Broyden's method (Q1184707)

This paper considers the problem of solving nonlinear systems of equations \(F(x)=0\), where \(F: \mathbb{R}^n\to \mathbb{R}^n\), \(F\) is differentiable and \(n\) is large. One of the most effective ways of solving the equations is \textit{C. G. Broyden's} ``good'' method [Math. Comput. 19, 577--593 (1965; Zbl 0131.13905)]. When the number of equations and unknowns is very large, memoryless implementation of this method are frequently used. The paper analyzes one of this implementations, and shows that calculations may be organized in such a way that parallelism can be exploited.