Performance analysis of shared buffer ATM switches with different cell departure models

Abstract

Performance analysis of an asynchronous transfer mode (ATM) switch involves modelling the input traffic source, the switching mechanism and the cell departure process. An important issue, which determines the overall accuracy in performance evaluation of the switch, is the use of an appropriate probabilistic model to describe the cell departure process. Until now, little work has been done in the literature on comparing models of the cell departure process for studying the performance of an ATM switch. This paper compares and assesses the accuracy of several models, including a new one proposed by us, called the 'urn model'. These models are put under test in a performance evaluation of a shared buffer ATM switch, by using a discrete-time Markov chain. The numerical results are compared with the simulation and show that the urn model is a good compromise between accuracy and efficiency. This finding is significant because it helps to speed up convergence of an analytical model of a large network while providing satisfactory accuracy. † This is referred to as a shared memory-type non-blocking switch, with all input and output ports being able to access a shared memory module (or shared buffer) at a maximum speed of reading or writing b cells in a clock cycle.

Queuing analysis of shared‐buffer ATM switches with grouped output channels

“…which is in good agreement with Equation (10) obtained by Fong and Singh in Reference [22]. The expressions we have derived in this section completely describe the departing probability "R +(x, X), (q, Q),.…”

An approximate transient analysis of dynamic thresholds in an ATM switch with shared buffer memory

“…where rn, k denotes the probability that k cells line up in n queues. Although we do not distinguish queue lengths among the lightly loaded ports, rn, k is approximately given as follows [10].…”