A ME-based approximation model for the performance evaluation of congestion control mechanism using queue thresholds

With the rapid development of the Internet, the control of congestion has become one of the most critical issues that must be confronted by the users. It is also a major challenge to the researchers in the field of performance modelling. This paper presents an approximate analytical performance model in a discrete-time queue, based on closed form expressions using queue threshold, to control the congestion caused by the bursty traffic. The methodology of maximum entropy (ME) is used to characterise closed form expressions for the state and blocking probabilities. A discrete finite capacity GGeo/GGeo/I/{N/sub 1/, N/sub 2/} censored queue with external compound Bernoulli traffic process and generalised geometric transmission times under a first come first serve (FCFS) rule and arrival first (AF) buffer management policy for single class jobs has been used for the solution process. To satisfy the low delay along with high throughput, a threshold, N/sub 1/ has been incorporated to slow the arrival process from mean arrival rate /spl lambda//sub 1/ to /spl lambda//sub 2/ once the mean queue length has been reached up to the threshold value N/sub 1/ (N/sub 2/ is the total capacity of the queue). The source operates normally, otherwise. This is like an implicit feedback from the queue to the arrival process. The system can be potentially used as a model for congestion control based on random early detection (RED) mechanism. Typical numerical results have been presented to show the credibility of ME solution and its validation against simulation.