Ad Hoc Distributed Simulation of Queueing Networks

Abstract

Ad hoc distributed simulation is an approach to predict future states of operational systems. It is based on embedding on-line simulations into a sensor network and adding communication and synchronization among the simulators. While prior work focused on this approach in the context of online management of transportation systems, this paper describes a generalization of the method and shows how it can be applied to embedded simulation of systems that can be modeled as a network of queues. An implementation of an ad hoc queueing network simulation is described. The flows of units across links connecting nodes in different simulations are approximated by renewal processes whose parameters are updated dynamically. The synchronization mechanism uses random sampling to update flow rates across simulations. Preliminary results show that the ad hoc queueing network simulation can provide predictions comparable to sequential simulations.