Modelling and Simulation of Queuing Models through the concept of Petri Nets

Abstract

In recent years Petri Nets has been in demand due to its visual depiction. Petri Nets are used as an effective method for portraying synchronization, a concurrency between different system activities. In queuing models Petri networks are used to represent distributed modeling of the system and thus evaluate their performance. By specifying suitable stochastic Petri Nets models, the authors concentrate on representing multi-class queuing systems of various queuing disciplines. The key idea is to define SPN models that simulate a given queue discipline 's behavior with some acceptable random choice. Authors have find system queuing with both a single server and multiple servers with load-dependent service rate. Petri networks in the queuing model have enhanced scalability by combining queuing and modeling power expressiveness of 'petri networks.' Examples of application of SPN models to performance evaluation of multiprocessor systems demonstrate the utility and effectiveness of this modeling method. In this paper, authors have made use of Stochastic Petri nets in queuing models to evaluate the performance of the system.