International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 15 - Number 2 |
Year of Publication: 2011 |
Authors: Md. Shohidul Islam, Md. Niaz Morshed, SK. Shariful Islam, Md. Mejbahul Azam |
10.5120/1920-2563 |
Md. Shohidul Islam, Md. Niaz Morshed, SK. Shariful Islam, Md. Mejbahul Azam . An Experimental Analysis of Random Early Discard (RED) Queue for Congestion Control. International Journal of Computer Applications. 15, 2 ( February 2011), 18-21. DOI=10.5120/1920-2563
Active Queue Management (AQM) is receiving wide attention as a promising technique to prevent and avoid congestion collapse in packet-switched networks. By providing advanced warning of incipient congestion, end nodes can respond to congestion before router buffer overflows and hence ensure improved performance. Random Early Discard (RED) is an IETF recommended active queue management scheme that is expected to provide several Internet performance advantages such as minimizing packet loss and router queuing delay, avoiding global synchronization of sources, guaranteeing high link utilization and fairness. It tends to drop packets from each connection in proportion to the transmission rate the flow has on the output link. It does not minimize the number of dropped packets as expected, but it manages to achieve improved performance when compared to the Tail Drop. In this paper, extensive experimental analysis has been carried out on RED using Network Simulator (NS-2) in relation to congestion control and decision has been settled where RED can perform better.