Optimistic Distributed Execution of Business Process Models

For the modeling of large and complex systems of business processes, a �ow oriented, graphical modeling framework based on Petri nets has emerged, taping the potentials of a qualitative and a quantitative analysis based on one and the same model. For the quantitative analysis of business process models �BPMs � representing realistically sized enterprise organizations, traditional evaluation techniques �like discrete event simulation � tend to become practically intractable. To be able to cope with very complex models, therefore, we have developed a distributed execution mechanism based on the Time Warp distributed simulation protocol. A corresponding software tool was implemented based on the MPI communication library, thus portable to almost any distributed orparallel computing platform. In case studies performed on a 134 node Meiko CS-2 multiprocessor investigating real and hypothetical business organizations, this work demonstrates that parallel�distributed simulation techniques make the execution of very large models feasible: in situations where the simulation model has reached a complexity prohibitive to an execution on a single processor computer �e.g. due to memory constraints�, the decomposition into smaller submodels to be executed on aparallel processor or a network of workstations remains the only means to get the simulation done. As such, a whole new class of �complex � BPM simulations becomes practically tractable, and traditional simulations can be accelerated dramatically. As an example, a BPM of a document �ow system comprising 64 o�ces with an average of 1000 documents per o�ce gains a 250 to 300 fold acceleration of the overall execution speed using 32 processors of the CS-2