Comparative analysis of queuing mechanisms: Droptail, RED and NLRED

Abstract

Congestion control in today’s Internet is an important issue. Congestion control algorithms have been extensively studied for managing the traffic and maintaining the stability in the network. In traffic management system, queuing plays an important role. This paper presents a comparison of three queuing mechanisms, namely Drop-tail, random early detection (RED) and nonlinear random early detection (NLRED) in wired network on the basis of different performance metrics such as end-to-end delay, throughput, packet drop and packet delivery ratio using NS2 simulator. The simulation results show that in high congestion, NLRED performs best while in low cohesive network Droptail gives good result. Also, we analyzed these queuing mechanisms in real audio traffic; again, all the experiments show that in congested network NLRED and RED are better while in low congested network Drop-tail is better because in heavy congested network congestion avoidance mechanism will help the network to achieve better performance. But in low congested network, the unnecessary computation avoidance mechanisms will degrade the network performance. However, if parameters are set effectively in RED, then it will be the best queuing mechanism for that particular network.