Swn analysis and simulation of large knockout ATM switches

Abstract

This paper approaches the performance evaluation of large ATM switches with Stochastic Well-formed Nets (SWN), a class of Colored Generalized Stochastic Petri Nets (CGSPN). The architecture of the ATM switches under investigation derives from the Knockout switch design, one of the most classical proposals for the implementation of large and fast ATM switches. The GSPN and SWN approaches to ATM network modeling are first discussed, then the Knockout architecture is presented, and the SWN models are illustrated. Results in terms of the state space complexity of the models and of the performance metrics obtained with different Knockout switch configurations are presented to prove the viability of the proposed approach.