摘要： 主动式队列管理技术是IETF为了解决Internet 拥塞控制问题而提出的一种路由器缓存管理技术。该文介绍了目前应用较为广泛的网络仿真器NS-2，对几种主要AQM算法Drop Tail、RED和ARED的性能在基于NS2仿真实验的基础上进行了比较研究，研究的性能包括队列长度、TCP 全局同步问题、连接数对系统稳定性和鲁棒性的影响等；仿真结果表明ARED性能优于Drop Tail和RED算法。

关键词: 网络仿真, 主动式队列管理, 尾丢弃, 自适应随机早期检测

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 words: Network simulation, Active queue management(AQM), Drop Tail, Adaptive random early detection(ARED)

中图分类号: 

• TP393