Using ODEs from PEPA models to derive asymptotic solutions for a class of closed queueing networks

In this paper a class of closed queueing network is modelled in the Markovian process algebra PEPA. It is shown that a fluid flow approximation using ordinary differential equations (ODEs) gives rise to well known asymptotic results. This result gives context to the use of a fluid flow approximation and is potentially useful in cases where the model is not obviously a closed queueing network. The approach is illustrated using examples of a secure key distribution centre and a multi-user query processing system. About the author Nigel Thomas joined the School of Computing Science in January 2004 from the University of Durham, where he had been a lecturer since 1998. His research interests lie in performance modelling, in particular Markov modelling through queueing theory and stochastic process algebra. Nigel was awarded a PhD in 1997 and an MSc in 1991, both from the University of Newcastle upon Tyne.Nigel is currently Director of Postgraduate Teaching within the School of Computing Science and Degree Programme Director for the MSc in Computing Science. Suggested keywords PERFORMANCE MODELLING, FLUID APPROXIMATION, STOCHASTIC PROCESS ALGEBRA Using ODEs from PEPA models to derive asymptotic solutions for a class of closed queueing networks Nigel Thomas ∗