Component‐Based Software Realisations of Monte Carlo and Stochastic Galerkin Methods

In this paper we present a component–based software realisation of the Monte Carlo and Stochastic Galerkin methods for simulating stochastic models with uncertain parameters described by random fields. The system supports parallel and distributed computations and focuses on the reusability and extensibility of the provided components and frameworks. Third–party software is interfaced to avoid cost–intensive reimplementations. The mentioned schemes are compared with respect to their convergence by considering a stationary groundwater flow model. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)