ScholarLens Back to search
2006•Jisuanji gongchengRequires access

Research of Active Queue Management Based on NS2

Weidong Zhang

Open publisher page 0 citations

Abstract

To address the problem of TCP end to end congestion control,IETF advocates deploying active queue management mechanisms in the TCP/IP networks.This paper introduces the widely used network simulator NS-2,several popular AQM algorithms,such as Drop Tail,random early detection(RED) and ARED (adaptive RED),are compared based on simulations.The performance metrics used in the study are queue size,TCP global synchronization problem,stability and robustness on different connections.The simulation results indicate that the performance of ARED is better than that of Drop Tail and RED.

About this research paper

What this paper is about

To address the problem of TCP end to end congestion control,IETF advocates deploying active queue management mechanisms in the TCP/IP networks.This paper introduces the widely used network simulator NS-2,several popular AQM algorithms,such as Drop Tail,random early detection(RED) and ARED (adaptive RED),are compared based on simulations.The performance metrics used in the study are queue size,TCP global synchronization problem,stability and robustness on different connections.The simulation results indicate that the performance of ARED is better than that of Drop Tail and RED.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

To address the problem of TCP end to end congestion control,IETF advocates deploying active queue management mechanisms in the TCP/IP networks.This paper introduces the widely used network simulator NS-2,several popular AQM algorithms,such as Drop Tail,random early detection(RED) and ARED (adaptive RED),are compared based on simulations.The performance metrics used in the study are queue size,TCP global synchronization problem,stability and robustness on different connections.The simulation results indicate that the performance of ARED is better than that of Drop Tail and RED.

Key concepts: Active queue management, Random early detection, Computer science, Queue, Network congestion, Robustness (evolution), Computer network, CUBIC TCP

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Research of Active Queue Management Based on NS2 — Research Paper | ScholarLens