输入排队中抢占式的短包优先调度算法

调度算法决定了输入排队交换结构的性能.本文根据Internet业务特征提出调度算法应保证短包的高优先级和低延迟.已有包方式调度中,长包信元的连续传输将造成短包长时间等待.为解决该问题,本文设计了一种低复杂度抢占式交换结构,并提出了相应的抢占式短包优先调度算法(P-SPF).短包优先可减小TCP流的RTT,并由此提高TCP之性能.通过排队论分析和实际业务源模型下仿真可知:P-SPF取得短包近似为零的平均包等待时间,同时达到94%的系统吞吐量.     

一种基于紧急程度的自时钟开始时间公平排队分组调度算法

为了克服目前GPS (Generalized Processor Sharing)类调度算法中实时应用分组的排队时延较大且不稳定的局限性,该文提出一种新的分组排队调度算法,该调度算法在计算分组服务标签时添加了一个紧急程度函数,调整了到达分组间的竞争关系,从而可以按照实时性应用的要求来调整到达分组的转发先优级,由此显著降低了实时性应用分组的排队时延和抖动幅度。分析和仿真实验表明,与GPS类其它调度算法相比,该调度算法对于实时应用的分组能提供较低的、更稳定的排队时延保证,同时还继承了GPS类算法的公平性和排队时延有界等特性,而且对系统虚拟时间的跟踪计算更为简捷高效。     

基于Crossbar的可重构网络输入排队分域调度研究

为解决传统网络技术体系中交换结构无法满足大量差异化业务规模化应用的问题,本文基于可重构网络技术体系,采用选择关闭部分Crossbar交叉节点的分域模型,提出了分域调度的思想,分析并推导了承载组内的SDRR调度算法和域内最长队列优先调度算法。最后采用交换性能仿真平台对该调度算法进行了复杂度和时延的仿真比较,结果表明:分域调度的最长队列优先算法比一般最长队列优先算法相对复杂度低,且随着调度域个数增加,相对复杂度降低。在相同业务源输入条件下,Crossbar三分域调度算法的时延小于非分域调度算法的时延,接近公平输出排队调度算法的时延。     

一种新的HSDPA分组调度算法研究

目前已提出的HSDPA调度算法仅是基于用户的优先级进行数据分组的调度.文章考虑了3G系统的业务多样性,综合考虑用户的优先级和业务的级别,创新性地提出将两类分组调度算法,即基于业务类别的分组调度算法--加权公平排队算法,与基于用户优先级的调度算法--最短队列算法或者最长队列算法相结合后的调度算法.仿真的结果显示,提出的调度算法能满足不同业务要求的QoS.从吞吐量和延时来看,最短队列算法与加权公平排队算法结合方案优于最长队列算法与加权公平排队算法结合的方案.     

一种基于时延补偿的星上分组调度算法

EDF策略具有有线调度算法中最大调度域和时延保障特性,根据分组时延要求确定服务优先级,是一种动态权重调度策略,通过将IQ-VOQ结构EDF策略引入无线调度并通过加入信道补偿机制和门限丢弃策略,得到一种新的算法IQCEDF,它避开公平排队固有的时延一带宽耦合问题,获得了比EDF更加出色的时延特性,更适于为卫星通信系统分组业务提供服务质量保证。     

多媒体无线网络中基于WRR的包调度策略研究

研究了多业务无线局域网中的分组调度策略,在WRR-CSDPS(信道状态相关的带权重的循环调度)算法的基础上提出了一种最长队列优先的调度策略。算法的目标是减小最长等待队列的长度。通过理论和仿真验证,在该算法下,系统的吞吐量和时延性能有所提高,该算法能够减小业务对系统缓冲区的容量要求。     

多用户MIMO-OFDMA系统混合业务的跨层分组调度算法

该文提出了一种针对多输入多输出-正交频分多址系统混合业务的跨层分组调度算法。该算法基于物理层的信道状况和媒体接入控制子层用户的排队时延,分别为实时业务用户和非实时业务用户的排队时延构建相应的效用函数,以保障不同业务的服务需求。仿真结果表明:该算法不但能保证实时业务的时延需要,而且能够获得很好的吞吐量,并满足公平性要求。     

iRGRR/PM：一种新的高速crossbar分组调度策略

iRGRR(iterative Request-Grant-based Round-Robin)算法是一种输入排队crossbar调度算法,具有硬件易实现、可扩展性强、性能优良等优点.在此基础上,该文提出一种新的高速crossbar分组调度策略iRGRR/PM(iRGRR with Packet Mode),可以更好地支持IP分组的调度,能够被应用于高速、大容量的路由器中.与iRGRR算法相比,iRGRR/PM简化了分组输出重组模块的设计,并且提高了crossbar的带宽资源利用率.文中简单分析了两种算法间的分组时延关系,并进行了详尽的仿真研究.结果表明:在相同条件下,iRGRR/PM算法具有更高的吞吐量,尤其在非均匀业务流下能获得接近100%的吞吐量;调度长分组时,iRGRR/PM算法具有更好的时延性能.     

具有优先服务机制的嵌套式DRR算法

针对DRR算法在保证业务时延性能上的固有不足,该文提出了具有优先服务策略的嵌套式DRR算法,该算法对需要保证延时的业务类实施专门的服务策略,利用漏桶控制和虚令牌的分配机制,并改善服务队列的调度策略,在不增加算法复杂度的情况下,有效地减小了时延敏感业务队列中包的延迟。文章证明了算法在获得延时保证的同时,选择合适的参数,仍然能够维持调度算法的公平性。仿真结果表明算法对时延性能的改善是明显的。     

基于队列长度和时延约束的异步OPS权重调度

分组调度在异步光分组交换(OPS)节点中起着重要作用,一种智能的调度算法能减小分组的丢失率,增加节点的吞吐量,从而提高服务质量.文章提出了一种基于队列长度和时延约束的解决队列中分组"饿死"和时延问题的权重调度算法.算法通过计算队列长度和分组在队列中等待时间确定调度的权重值,以减小分组在核心节点处的资源竞争和等待时延,调度器计算每个输入端口的N×K个虚拟输出队列的权重值并调度最大权重值队列到可用的输出信道上.分析和仿真结果表明所提的权重调度算法具有高吞吐量和低分组丢失率特点.     

Optimal Packet Length for Packet Voice Communication

It is shown that for typical operating parameters, the optimal packet length for a single link packet voice system is of the order of 300-700 bits. This is contrary to both the optimal length of a data packet (approximately 1000 bits), and documented experimental implementations of such network architecture for voice (approximately 1000 bits).     

Packet by Packet Analysis in Contemporary Network Simulators

This article presents a comparative study of three hybrid network simulators: OPNET’s Modeler and ns-2, as well as NCTUns. In order to fulfill this, the simulated results were compared with the results obtained from a real network testbed using a packet by packet analysis methodology. Using a generic network based in an experimental network and modeled in the three simulators, the experimental comparison consisted in introducing data traffic in six different scenarios, combining traffic patterns of Constant Bit Rate (CBR) and a File Transfer Protocol session (FTP). The first objective is to corraborate the precision of the used models in the simulation to create this traffic patterns. Second, to test the network response to these types of traffic, all these by using the same packet by packet analysis methodology. The results obtained from the simulators were compared against the results obtained from the testbed network. The results obtained in this study showed that ns-2 presents good results in the simulation of CBR traffic, while the Modeler from OPNET presents an outstanding traffic simulation in terms of the FTP traffic. Nevertheless, some problems were detected in the three simulators, but NCTUns was the one that presented the most number of errors and problems in the overall results. These types of studies at the packet by packet level are important to validate the models used in the simulations that are based in components and applications used in grand scale simulations.     

走出IP网络的“迷宫”

11月17日,美国百科帝(Packet Design)公司的首席技术官Cengiz Alaettinoglu博士到中国访问。Cengiz博士在管理广域网的路由分析技术领域享有知名地位,曾任互联网工程任务组(IETF)路由政策系     

Packet video

The 1990s will see a large increase in the use of video for many purposes, made possible through the use of new optical storage and transmission methods, as well as radio and satellite communication. HDTV, new broadcast services and 64 kbit/s ISDN video are all at an advanced stage of development. Digital methods of transmission and storage feature prominently in future scenarios of video usage. To avoid long delays in transmission, packet video techniques for storage of high-quality material on optical discs are being considered. Coded video signals can be transmitted in packet form at variable bit rate. The author discusses the principles of packet video. Packet loss, queueing delay and layered coding are discussed. The future developments are also discussed     

Packet forwarding

A packet network is viewed as consisting of two major interacting layers: a lower layer responsible for the determination of a set of paths that can be used to carry packet flows, and an upper layer responsible for actually sending the flows over these paths, on a per-packet basis. A systematic review is provided of the various techniques that can be used to perform the actual packet-by-packet routing operation at the upper layer, called the packet forwarding function. The function components of a routing system and its characteristics (responsiveness to changes in network state and degree of centralization) are discussed. Five routing techniques are then examined, namely source, directory, destination, global-path, and channel-link-path routing     

Packet phone

Why would anyone want to transmit voice over the creaky, packet-based Internet? In this paper, the author looks for some answers to the great conundrum of modern communications     

Packet Pacing in Small Buffer Optical Packet Switched Networks

In the absence of a cost-effective technology for storing optical signals, emerging optical packet switched (OPS) networks are expected to have severely limited buffering capability. To mitigate the performance degradation resulting from small buffers, this paper proposes that optical edge nodes “pace” the injection of traffic into the OPS core. Our contributions relating to pacing in OPS networks are three-fold: first, we develop real-time pacing algorithms of poly-logarithmic complexity that are feasible for practical implementation in emerging high-speed OPS networks. Second, we provide an analytical quantification of the benefits of pacing in reducing traffic burstiness and traffic loss at a link with very small buffers. Third, we show via simulations of realistic network topologies that pacing can significantly reduce network losses at the expense of a small and bounded increase in end-to-end delay for real-time traffic flows. We argue that the loss-delay tradeoff mechanism provided by pacing can be instrumental in overcoming the performance hurdle arising from the scarcity of buffers in OPS networks.      

分组网络中包调度算法研究

下一代因特网要支持QoS(quality of service)实现，而包调度又是QoS实现的关键技术之一。文中对现有的包调度算法进行了分类，分析和比较了各类算法的技术特点与性能优劣，讨论了包调度算法研究今后的发展方向。     

A Scalable Packet Sorting Circuit for High-Speed WFQ Packet Scheduling

A novel implementation of a tag sorting circuit for a weighted fair queueing (WFQ) enabled Internet protocol (IP) packet scheduler is presented. The design consists of a search tree, matching circuitry, and a custom memory layout. It is implemented using 130-nm silicon technology and supports quality of service (QoS) on networks at line speeds of 40 Gb/s, enabling next generation IP services to be deployed.     

Round-Robin分配方式下多径传输的重排序性能分析

基于包分配的多径传输在接收端所引起的数据包乱序严重影响TCP的传输性能。针对此问题,从排队论的角度出发对Round-Robin分配方式下多径传输的重排序问题进行了分析。从数据包的乱序率、重排序时间和端到端总时间3个方面考察了路径差异与路径数目对多径传输的性能的影响。结果表明,在采用两条路径传输时,应使得两条路径的传输速率近似相同;在路径传输速率相同的条件下,为明显地提升多径传输的重排序性能,路径数目应不超过4条。