Fast cell loss rate estimation of ATM switches using importance sampling

The performance evaluation of ATM switches is of paramount importance in designing an ATM network. We focus on the evaluation of the cell loss rate (CLR) in nonblocking ATM switches using computer simulations. In particular, we investigate the potential of using importance sampling techniques as a “superfast” alternative to conventional Monte Carlo simulation in finding the CLR in nonblocking ATM switches. We propose a “split switch” method to decouple the input and output queue behaviour, along with the notion of regenerative cycles to achieve fast and accurate results. Numerical results demonstrate that a considerable computation cost can be saved using these proposed importance sampling techniques while maintaining a high degree of accuracy