A Parallel Modified Lagrangian Method for an Optimal Control Problem of a Linear Distributed Stochastic System

We consider an optimal control problem for a Hilbert space valued linear stochastic evolution equation with additive noise and a quadratic goal function in integral form. We solve an innnite dimensional linear stochastic regulator problem by applying an augmented Lagrangian algorithm. Because of high amounts of computation time required by numerical solution of such problems we investigate a parallelization of this algorithm by distributing the computations of space stages over several processor nodes of a parallel computer. We obtain an eecient parallel algorithm with low communication cost by using a ring topology.