Modelling and analysis of parallel information systems.

This thesis presents an investigation of modelling and analysis of parallel information systems. The research was motivated by the recent developments in networks and powerful, low-cost, desk top multiprocessors. An integrated approach for the construction of parallel information systems was developed which focussed on modelling, verification and simulation of such systems. The thesis demonstrates how Petri nets can be used for the modelling and analysis of entity life histories and parallel information systems, place transition nets for the modelling and analysis of entity life histories and coloured Petri nets for the modelling and analysis of complex parallel information systems. These tools were integrated into a comprehensive framework which allowed for the modelling and analysis of complex parallel information systems and the framework was tested using a comprehensive case study. The thesis concludes that Petri nets are an ideal tool for the modelling and analysis of complex parallel systems. Verification is possible with deadlocks and similar properties being easily identified. Further the transformation rules proved to be beneficial to the process of moving from one model to another. Finally simulation of parallel behaviour was possible because the underlying models captured the notion of parallelism.