Title :
Modelling active queue management with different traffic classes
Author :
Wang, Lan ; Min, Geyong ; Awan, Irfan
Author_Institution :
Dept. of Comput., Bradford Univ., UK
Abstract :
Random early detection (RED), one of the most popular active queue management (AQM) schemes, has become an effective mechanism for supporting end-to-end traffic congestion control and has attracted lots of research efforts on the analysis of its performance properties under various networking scenarios using software simulation. This paper develops an analytical performance model for a finite capacity queueing system with AQM mechanisms based on the instantaneous number of packets in the queue. We consider two different traffic classes modeled by Poisson arrival processes. Expressions for the aggregated and marginal performance measures including the mean queueing length, response time, system throughput, probability of packet losses and mean waiting time are derived. The validity of the analytical results is demonstrated by comparing the analytical results and those obtained from simulation experiments.
Keywords :
probability; queueing theory; stochastic processes; telecommunication congestion control; telecommunication network management; telecommunication traffic; AQM scheme; Poisson arrival process; RED; active queue management; analytical performance model; end-to-end traffic congestion control; finite capacity queueing system; probability; random early detection; software simulation; Analytical models; Communication system traffic control; Length measurement; Mechanical factors; Performance analysis; Performance loss; Queueing analysis; Software performance; Telecommunication traffic; Traffic control;
Conference_Titel :
Advanced Information Networking and Applications, 2006. AINA 2006. 20th International Conference on
Print_ISBN :
0-7695-2466-4
DOI :
10.1109/AINA.2006.233
