DEMAND AND SERVICE MATCHING AT HEAVY LOADS - A DYNAMIC BANDWIDTH CONTROL MECHANISM FOR DQDB MANS

In this paper, the Alpha Tuning Mechanism for distributed queue dual b us (DQDB) networks is proposed. This mechanism achieves a match betwee n the users' demand and the resulting protocol's service patterns. The use of a tunable alpha-parameter (representing some fictitious user p opulation) and access protection lead to dynamic bandwidth control cap ability. The inherently position dependent fairness problems of DQDB a re overcome, especially at heavy loads. Fairly uniform delays and unif orm success rates of transmission are seen for every user. Analysis an d simulation studies for different load patterns, show that it outperf orms the bandwidth balancing mechanism (BBM). This mechanism may be ex tended to any unidirectional channel, to achieve demand and service pa ttern match.