Quantitative Adaptive RED in Differentiated Service Networks

This paper derives a quantitative model between RED (Random EarlyDetection) max_p andcommitted traffic rate for token-based marking schemes inDiffServ IP networks. Then, a DiffServ Quantitative RED (DQRED) ispresented, which can adapt its dropping probabilityto marking probability of the edge router to reflect not only thesharing bandwidth but also the requirement of performance of theseservices. Hence, DQRED can cooperate with marking schemes to guaranteefairness between different DiffServ AF class services. A new markingprobability metering algorithm is also proposed to cooperate with DQRED.Simulation results verify that DQRED mechanism can not only controlcongestion of DiffServ network very well, but also satisfy differentquality requirements of AF class service. The performance of DQRED isbetter than that of WRED.