Skip to content. | Skip to navigation

Personal tools

Roland Ewald (2006)

Simulation of load balancing algorithms for discrete event simulations

Diplomarbeit, Universität Rostock.

Efficiently simulating discrete-event models in a parallel and distributed manner is a challenging endeavour. On one hand, various factors, such as hardware infrastructure or model characteristics, have to be considered. On the other hand, there is a wide variety of algorithms which address ubproblems of parallel and distributed simulation and whose performance depends on the application at hand. This work illustrates the resulting difficulties with respect to the development of parallel and distributed simulation systems for general purposes. It is reasoned to what extent a prior analysis could facilitate this task. Approaches to predict the performance of parallel and distributed simulation systems are discussed. It is argued that the simulation of such a simulation system is a feasible approach. This claim is underpinned by implementing SimSim, a sequential simulator to simulate parallel and distributed simulation systems. Then, SimSim is used to develop a load balancing algorithm for the simulation of PDEVS models. The algorithm’s performance is analysed using SimSim. Finally, the predicted performance is compared to the real performance of the algorithm when running in the simulation system James II.