TCP – Random Early Detection (RED) mechanism for Congestion Control

This thesis discusses the Random Early Detection (RED) algorithm, proposed by Sally Floyd, used for congestion avoidance in computer networking, how existing algorithms compare to this approach and the configuration and implementation of the Weighted Random Early Detection (WRED) variation. RED uses a probability approach in order to calculate the probability that a packet will be dropped before periods of high congestion, relative to the minimum and maximum queue threshold, average queue length, packet size and the number of packets since the last drop. The motivation for this thesis has been the high QoS provided to current delay-sensitive applications such as Voice-over-IP (VoIP) by the incorporation of congestion avoidance algorithms derived from the original RED design [45]. The WRED variation of RED is not directly invoked on the VoIP class because congestion avoidance mechanisms are not configured for voice queues. WRED is instead used to prioritize other traffic classes in order to avoid congestion to provide and guarantee high quality of service for voice traffic [43][44]. The most notable simulations performed for the RED algorithm in comparison to the Tail Drop (TD) and Random Drop (RD) algorithms have been detailed in order to show that RED is much more advantageous in terms of congestion control in a network. The WRED, Flow RED (FRED) and Adaptive RED (ARED) variations of the RED algorithm have been detailed with emphasis on WRED. Details of the concepts of forwarding classes, output queues, traffic policies, traffic classes, class maps, schedulers, scheduler maps, and DSCP classification shows that the WRED feature is easily configurable on tier-1 vendor routers