Title :
Simulation and evaluation of random early detection in congestion control
Author_Institution :
Coll. of Comput. Sci. & Technol., Yantai Univ., Yantai
Abstract :
Congestion control maintains the number of packets within the network below a certain level to avoid low performance. And queue management is an important and critical component of any network implementation. RED is proposed to overcome some shortcomings of traditional methods for queue management and is proved to be more efficient than drop tail. In this paper we design a network environment using RED protocol and simulate it with NS-2. From the analysis of simulation results we find that the RED proactive discarding can provide low delay and proportional dropping fairness. On the other hand, when RED is used, more packets are lost and the throughput is a bit lower but very close to that when drop tail is used. To conclude, RED can offer high QoS in some aspects.
Keywords :
quality of service; queueing theory; telecommunication congestion control; telecommunication network management; QoS; RED protocol; congestion control; network environment; proportional dropping fairness; queue management; random early detection; Communication system traffic control; Cybernetics; Delay; Machine learning; Protocols; Quality of service; Switches; Tail; Throughput; Traffic control; Congestion control; Drop tail; NS-2; QoSRED; Simulation;
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620800