COMPARATIVE PERFORMANCE OF SCHEDULING STRATEGIES FOR SWITCHING AND MULTIPLEXING IN A HUB BASED ATM NETWORK - A SIMULATION STUDY

We model an ATM network comprising an input quelling cell switching hu b. Customer access lines are multiplexed into the relatively faster in put links of this hub. Each access multiplexer is fed by on/off source s. Motivated by the fact that an ATM network should support both burst y as well as smooth traffic, we consider the scenario in which some mu ltiplexers are fed by sources with long bursts of cells, and others by sources with short bursts. Here we report the results of a detailed s imulation of this hub-based ATM network. Our objective is to compare t he performance of various strategies for scheduling cell service in th e access multiplexers, and in the ATM switch. The simulation results c onfirm what might be expected from the results of our earlier analytic al modelling of the multiplexer and the switch in isolation (this anal ysis assumed a particular Markovian model for the aggregate cell arriv al processes into the ATM switch). In particular, we find that, if mea n burst delay is the performance criterion then, for a small ratio of ATM-Link to customer-access-line speed (less than or equal to 3 for th e models and parameters we use), the more bursty traffic should underg o burst level multiplexing at the access multiplexer, and should be gi ven lower priority during output contention resolution in the input qu eueing hub. If worst case delay performance (e.g., the 99.9 percentile of the end-to-end delay of a burst) is a consideration, however, we f ind that, even for small ATM-link to access-line speed ratio, the best combination of strategies is that the access multiplexers should mult iplex the packets from the access lines in a round-robin fashion, and the switch should still give lower priority to the burstier traffic. ( C) 1998 ElsevierScience B.V. All rights reserved.