Performance of Active Queue Management Algorithms to Be Used in Intserv Under TCP and UDP Traffic

摘要

Active Queue Management techniques are recommended to overcome the performance limitations of TCP congestion control mechanisms over drop-tail networks. Flow Random Early Drop (FRED), GREEN, Stochastic Fair Blue (SFB), Stabilized RED (SRED) are some of the active queue management algorithms which are flow-based in nature. These algorithms can be used to implement Intserv. The main objective of this paper is to present a comparative analysis of the performance of the FRED, GREEN, SFB, and SRED algorithms using the NS-2 network simulator. In this work, the simulations are carried out for the comprehensive analysis and comparison of the algorithms. The algorithms are tested in terms of average queue size, fairness, utilization and packet loss rate by applying various number of flows under TCP, and TCP/UDP traffic.