SBio:一种基于DiffServ网络的主动式队列管理算法  

SBio: a new AQM algorithm for DiffServ network

作　　者：吴春明[1] 姜明[2] 

机构地区：[1]浙江大学人工智能研究所,浙江杭州310027 [2]杭州电子科技大学计算机学院,浙江杭州310018

出　　处：《通信学报》2005年第6期130-136,共7页Journal on Communications

摘　　要：区分服务体系结构由于其实现简单、扩展性好,目前已经成为解决IPQoS问题的主要模型。区分服务网络中核心路由器执行的是RIO算法,以对不同优先级的数据包进行管理。但是RIO算法不能有效地保持队列长度的稳定性,因此会导致不可预测的排队延迟和延迟抖动,另外RIO算法还存在“带宽偏移”问题。SBio算法的目的是维持队列长度的稳定并且减轻“带宽偏移”问题。SBio算法是在SBlue算法基础上进行的扩充,使其能管理不同优先级的队列。仿真实验表明,与RIO算法相比,SBio算法能较好地维持队列长度的稳定并有效地减轻“带宽偏移”问题。The DiffServ architecture has recently become the preferred service model for addressing QoS issues in IP networks due to its simplicity and scalability. In DiffServ domain, the core routers use RIO algorithm, which performs different packets dropping schemes when congestion occurs. But RIO algorithm cannot stabilize the queue size and therefore, leads to unpredictable queuing delay and jitter. Furthermore, RIO has the “bandwidth skew” problem. SBio was a new AQM algorithm for DiffServ network, which runs two different SBlue algorithms for IN and OUT packets. The object of SBio was to stabilize the queue size and mitigate the “bandwidth skew” problem. The simulation results indicate that compared with RIO, MRIO performs better in stabilizing queue size and mitigating “bandwidth skew” problem.

关 键 词：区分服务 RIO SBlue SBio 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]