RED改进技术概述

RED技术的产生为解决网络的拥塞提供了良好的办法,但随着网络的发展、业务的丰富,RED技术的缺点也越来越明显。这种缺点的显现也推动了一系列的改进算法的产生,本文概述了各种改进算法之间的关系和各种算法的技术特点。     

RED的网络性能研究*

在很多网络中, 支持新的拥塞避免机制( 如RED 和ECN) 的节点将会和支持传统丢尾缓存( Drop-Tail) 的节点共存。研究含有两个拥塞节点的混杂网络中TCP 流集的性能, 比较了几种控制算法的数据流公平性, 仿真结果显示采用Adaptive 算法并不能增加数据流的公平性; 仿真也显示混杂网络中单个RED 参数的设置情况会影响整个网络的性能, 具体可以从网络吞吐量、响应时间等判断。RED 的参数设置在双瓶颈情况下再次变得重要, 正如在单瓶颈的情况下一样。     

图书馆多媒体信息资源网络的自适应传输控制方法

为了增强图书馆多媒体信息的稳健传输和多媒体信息网络的鲁棒性、稳定性,提出了一种新颖的RED策略。最小阈值(minth)和最大阈值(maxth)能够随着缓存队列的平均占用率的变化动态调整,而不再是预设的固定参数。改进后,ATRED对路由器队列的控制力度能够根据网络环境的变化自适应调整,使得队长方差减小,队列振荡变小,网络更加稳定,能够在复杂多变的网络环境下提供更加可靠的服务。     

Agro-ecosystem modeling can aid in the optimization of biomass feedstock supply

Recent European Directives promoted the development of biofuels, requesting mandatory limits to their emissions ot greenhouse gases (GHG). Second-generation biofuels based on lignocellulosic biomass are prime candidates but their GHG emissions are variable and uncertain. Agro-ecosystem modeling can capture them and the performance of biofuel feedstocks.This study aimed at optimizing feedstock supply for a bioethanol unit in France, from agricultural residues, annual and perennial crops. Their productivity and environmental impacts were modelled on a regional scale using geo-referenced data on soil properties, crop management, land-use and future weather data. Several supply scenarios were tested. Cereal straw was the most efficient feedstock but had a low availability, and only miscanthus could meet the bioethanol plant's demand. Sorghum combined poor yields and high GHG emissions compared by miscanthus and triticale. A mix of three biomass sources used less than 3% of the regional agricultural land while abating GHG emissions by 60%.     

基于多优先级的动态阈值RED算法

提出一种基于多优先级的动态阈值RED算法(PDT-RED)。根据分组优先级和未使用的缓存空间动态调节平均队列阈值,实现对不同类型的分组采取不同的标记/丢弃概率的区分服务。为了提高缓存利用率,在该算法基础上提出优化算法(OPDT-RED)。模拟实验证明,与同类算法相比,OPDT-RED使高优先级的分组丢失率降低了56%~62%,平均分组丢失率降低了33%~41%,并且简单、易于实现,能提高缓存的利用率。     

改进的TCP/RED模型的稳定性分析

TCP自同步特性可提高网络稳定性,TCP/RED拥塞控制模型忽略了TCP自同步特性。针对该问题,提出包含TCP自同步特性更接近实际网络的改进TCP/RED拥塞控制模型。应用Nyquist稳定性判据给出当改进模型在复平面上属于某一不包含-1+j0点的下半平面时的稳定性条件,得到比TCP/RED模型更宽松的RED参数设置范围。NS2仿真验证了该稳定性条件的有效性。     

Performance evaluation for DRED discrete-time queueing network analytical model

Due to the rapid development in computer networks, congestion becomes a critical issue. Congestion usually occurs when the connection demands on network resources, i.e. buffer spaces, exceed the available ones. We propose in this paper a new discrete-time queueing network analytical model based on dynamic random early drop (DRED) algorithm to control the congestion in early stages. We apply our analytical model on two-queue nodes queueing network. Furthermore, we compare between the proposed analytical model and three known active queue management (AQM) algorithms, including DRED, random early detection (RED) and adaptive RED, in order to figure out which of them offers better quality of service (QoS). We also experimentally compare the queue nodes of the proposed analytical model and the three AQM methods in terms of different performance measures, including, average queue length, average queueing delay, throughput, packet loss probability, etc., aiming to determine the queue node that offers better performance.     

一种改进AVQ的主动队列算法

在讨论了各种算法的基础上，分析和比较了现有的几种主动队列管理算法：随机早期检测算法（RED），自适应RED算法（ARED），自适应虚拟队列（AVQ）算法，动态阈值（DT）算法以及队列长度阈值（QLT）的分组调度算法等，并对AVQ算法进行了改进，将原算法维持单个虚拟队列改为维持多个虚拟队列，从而使其在原性能的基础上增加了区分服务的功能，以达到适应下一代网络发展的需要。     

随机早期诊断(RED)队列管理算法的改进研究

本文总结了近十年来RED队列管理机制的各种改进研究思路，并对限制高带度流量进行了仿真以证实其对丢包性能的影响，最后分析了根据总的丢包量度大小来估计某流量的到达率的不合理性，从而我们很有必要利用仿真工具来验证比分析的正确性，并对RED加以修正使其拥塞控制的性能更好。     

路由器QoS模块研究

互联网从研究网络到商业网络的转化以及多种对QOS有不同要求的应用的出现要求互联网提供QOS．本文首先论证了在路由器中实现QOS的必要性，然后以此为基础，给出了在路由器中实现QOS模块所需要考虑的问题和实现框架，并详细讨论了实现细节．最后在具有自主知识产权的BQ2000路由器上对该模块作了测试，给出了评测．