光纤通道交换机在强实时约束下的分组调度

以光纤通道交换网络强实时约束下的性能研究为背景,采用实时通信中的周期性任务模型,提出了负载匹配的加权轮循分组调度,导出了在该方法下网络消息集严格实时的充要条件,以最差情形下强实时的网络可达负载率为性能衡量指标推证了采用该算法的优越性并通过仿真进行了验证.     

EPON中基于滑动周期的动态带宽分配改进算法

动态带宽分配(DBA)算法是EPON技术的关键.传统的间插轮询自适应周期长度(IPACT)算法被证明在不规则流量下会遭遇严重的带宽利用率下降.基于滑动周期的SLICT算法可以解决这一问题,然而公平收敛性和带宽利用率之间的冲突限制了其在高突发性网络环境下的性能.本文提出一种改进的SLICT算法,在每次授权时都综合考虑网络实时负载变化.仿真结果证明改进的SLICT算法可以同时达到高带宽利用率和有保证的公平收敛性,性能较现有算法有进一步提高.     

基于优先级控制的NCS数据传输性能研究

NCS对实时性数据传输性能要求的提高使得传统WFQ算法无法保证实时业务QoS，本文，指出了WFQ算法的公平调度原则存在的不足，提出了一种基于优先级分配带宽的调度算：PLLQ。仿真结果显示，PLLQ算法不仅改善了强实时业务的延时性能，并且解决了非实时性业务延时过长的缺点，提高了NCS的数据传输性能。     

实时计算机网络的有效带宽

文章通过剖析实时计算机网络研究中“有效带宽”定义的内涵与外延,阐明了它在介质访问仲裁、逻辑链路控制、确定性排队、保证概率排队等应用领域中的意义与用法;并指出了根据大偏离原理(LargeDeviationPrinciple,LDP)定义的有效带宽不仅从概念上反映了多时间尺度和空间尺度下网络的统计复用情况,而且具有带宽的量纲。通过有效带宽与流量过程的累积量生成函数(CumulantGeneratingFunction,CGF)的关系,给出了将大偏离原理应用于流量整形变长数据包流量的方法,并根据随机序(stochasticordering)理论提出了一种“4状态7参数”模型作为这种整形数据流的随机性能界限。通过对CGF函数在空间尺度上求极限,得到最大流率函数,揭示了LDP框架下,保证概率(guaranteeingprobability)意义下到最坏条件下(theworstcase)实时性能保证具有一致的语义。     

航空电子FC不同拓扑和服务类的实时性测试

在不同的消息流分布模式下,对光纤通道（FC）各种拓扑结构和服务类进行了性能测试,给出了各种传输情况下的消息延迟和吞吐量随负载的变化关系。通过分析得出结论：光纤通道的交换网络和服务类3分别较仲裁环和服务类2延迟小,最大吞吐量大;FC网络在均匀负载模式下,累积负载为网络最大吞吐量的50%以下时,消息传输具有确定性延迟上限。     

基于多蚁群算法的无线传感器网络路由的跨层设计

针对事件驱动型无线传感器网络对数据传输的强实时性和高可靠性的要求,将统计获得的节点接入效率和负载队列长度等MAC层状态信息作为路由度量参数。考虑到蚂蚁网络算法在重负荷网络中存在传输延迟大、聚合速度慢等缺陷,提出了一种基于多蚁群算法的负载感知和高效接入的跨层路由协议,该协议把节点的单跳延迟、负载及带宽接入效率等参数作为路径的启发值进行路由优化。仿真结果表明,基于多蚁群算法的路由协议能够保证数据传输的实时性,能够实现无线传感器网络的拥塞控制及负载平衡。     

一种FlexRay总线优化方法研究

为了提高FlexRay车载网络带宽利用率,减小消息帧的最坏响应时间,对FlexRay总线配置问题进行研究。针对FlexRay协议中静态段负载长度相等,静态消息的长度不等的情况,找到最优静态段负载长度,提高带宽利用率。基于静态段的优化之上,综合分析静态段、动态段消息的最坏响应时间,得到最优总线周期长度。实验结果表明,在得到最优负载的条件下,带宽利用率提高22%,并且在周期长度为982μs时,总线消息帧的最坏响应时间最小,保证了消息传输的实时性、可靠性。     

基于移动网络的远程视频监控系统*

针对将远程监控扩展到移动网络时存在的移动设备计算和显示能力、移动网络数据传输能力的瓶颈,提出了使用MPEG-4视频压缩结合尺寸压缩的数据压缩方案,充分降低了网络数据流量,减轻了移动终端的渲染计算压力;提出了考虑移动设备计算能力的基于反馈机制的自适应网络传输策略以解决累积延时效应,在数据传输速度与变化的移动网络带宽和移动设备性能之间构建动态平衡,保证移动监控的实时性。通过原型系统的测试,证明该方案可充分降低网络流量、保证监控的实时性,同时具有一定的普适性。     

无线Mesh网络中VOIP的QoS性能分析

针对VoIP(Voice over IP)业务在无线Mesh网上进行传输时存在服务质量(QoS)需求难以保证、带宽利用率低的问题,介绍了VoIP的QoS影响因素,分析了端到端时延、时延抖动和丢包率等几个重要参数,并对VoIP在无线Mesh网中的传输性能进行了论述。提出了基于无线Mesh网络的QoS保证机制,可以为端到端的数据传输公平的分配带宽,并能在保证QoS下实现大规模的实时任务的多跳转发。仿真试验表明能有效降低端到端时延,有着更好的QoS性能。     

基于GIS的分布式实时协同制图系统的研究

由于Internet网络带宽限制等诸多方面的影响,基于GIS的分布式实时协同制图系统是较难实现的。采用复制式模型构造了一个基于GIS的分布式实时协同制图系统,并且采用动态数据格式作为各站点的交换数据,极大限度地战少了网络数据传输量,提高了系统的响应速度和稳定性,满足了协同制图系统对实时性的要求。     

Message transport on the Cambridge ring—A simulation study

This paper describes a message transport policy for the Cambridge Ring which introduces contention for the receiver. A simulation model is presented which makes worst case assumptions for transmission rates. The results of running this simulation model are presented in terms of the delays experienced by messages crossing the ring given a certain discrete distribution of message sizes. Three variations of the policy are compared under different loadings of the ring caused by the number of nodes using the ring and by message creation rates at each node.     

COTS Switch工业以太网实时调度算法

针对基于商用现成组件技术（Commercial Off-The-Shelf,COTS）的交换式以太网不能满足工业数据通信的实时性要求问题,给出了改进的FTT（Flexible Time-Triggered communication paradigm）网络调度模型,提出了新的周期性实时消息链路可调度性优化判定方法,在此基础上设计了一种基于EDF（Earliest Deadline First）的实时调度算法。算法通过对网络消息流量进行有效控制防止交换机缓冲区溢出,同时充分发挥网络在同一时间存在多个并行传输链路的优势,提高了网络实时性。仿真实验表明该算法在提高网络的聚合带宽、减小消息的平均等待延时和丢失率方面均有明显的优势。     

Fault-aware Communication Mapping for NoCs with Guaranteed Latency

As feature sizes shrink, transient failures of on-chip network links become a critical problem. At the same time, many applications require guarantees on both message arrival probability and response time. We address the problem of transient link failures by means of temporally and spatially redundant transmission of messages, such that designer-imposed message arrival probabilities are guaranteed. Response time minimisation is achieved by a heuristic that statically assigns multiple copies of each message to network links, intelligently combining temporal and spatial redundancy. Concerns regarding energy consumption are addressed in two ways. First, we reduce the total amount of transmitted messages, and, second, we minimise the application response time such that the resulted time slack can be exploited for energy savings through voltage reduction. The advantages of the proposed approach are guaranteed message arrival probability and guaranteed worst case application response time.     

Systems of recall broadcast

In a system of recall broadcast, each message contains a set of message identifiers called its recall set. The recall set of a message identifies all previously received messages upon which the message is based. When a message is received by a process and displayed to an external user, all messages in its recall set are also displayed to the user. Our recall broadcast system is efficient in three ways. First, each message is broadcasted only once. Second, each process stores a small number of previously received messages because these messages may be recalled in future messages. Third, processes exchange additional messages to delete previously received messages that can no longer be recalled. The additional messages have the same format as regular messages, and are handled exactly as regular messages.     

一种IMS网络信令异常处理检测方法

针对IMS网络中被入侵或劫持的网络实体可能对信令消息做出恶意篡改等异常处理的问题,提出一种基于信令处理规则的IMS 网络信令异常处理检测方法.该方法基于建立的信令处理规则库,模拟实体对信令消息的正常处理并生成一条预处理消息,通过判断预处理消息与经过实体处理后的信令消息是否匹配,检测IMS网络中是否存在信令异常处理.实验结果表明,该方法对信令异常处理的检测率达到了100％.     

A Token-Based Distributed Group Mutual Exclusion Algorithm with Quorums

The group mutual exclusion problem is a generalization of mutual exclusion problem such that a set of processes in the same group can enter critical section simultaneously. In this paper, we propose a distributed algorithm for the group mutual exclusion problem in asynchronous message passing distributed systems. Our algorithm is based on tokens, and a process that obtains a token can enter critical section. For reducing message complexity, it uses coterie as a communication structure when a process sends a request messages. Informally, coterie is a set of quorums, each of which is a subset of the process set, and any two quorums share at least one process. The message complexity of our algorithm is $O(|Q|)$ in the worst case, where $|Q|$ is a quorum size that the algorithm adopts. Performance of the proposed algorithm is presented by analysis and discrete event simulation. Especially, the proposed algorithm achieves high concurrency, which is a performance measure for the number of processes that can be in critical section simultaneously.     

DTN中基于节点价值的效用路由算法

为了解决延迟容忍网络(DTN)中传统路由算法中消息被分配的网络资源不均衡及节点负载不均衡问题,结合消息效用值提出了一种基于节点价值的效用路由算法。算法根据动态改变的消息效用值选择最高优先级的消息(具有最小TTL和到目的节点最短距离的消息)进行转发,以使得为每个消息分配的网络资源相对均衡;同时,根据节点的价值(与节点速度和剩余缓存有关)选择下一跳节点,以平衡每个节点的负载;另外,算法还采用了一定的消息管理机制及时清除缓存空间。通过仿真实验及性能分析表明,该算法在传输成功率、传输延迟和网络开销上都有明显的改善。因此,通过充分利用网络资源提高了算法的整体性能。     

Optimal scheduling with strict deadlines

The problem of dynamic scheduling of customers (messages) in time-critical environments is discussed. A single station (communication node) is considered, and it is assumed that each customer (message) must begin service (transmission) by an individually varying extinction time or else it is lost. Interest is in minimizing, in the sense of stochastic order, the number of messages lost over any time interval. A variety of results is proved that establishes the optimality of the shortest-time-to-extinction policy under rather general conditions. Similar results are found when messages have constraints on their complete transmission times. A network of M stations in tandem is considered under the hypothesis that a message is never lost and is scheduled irrespective of whether its extinction time (also called due date in this case) has expired or not. Under fairly general assumptions on the arrivals, deadlines, and services, it is shown that the earliest-due-date policy minimizes a form of average tardiness incurred over a finite operating horizon among all non-idling nonpreemptive policies. These problems are formulated in the context of stochastic dominance, and simple interchange arguments are used to establish all results     

基于发布/订阅模型的异构网络数据交换方法

消息数据高效传输是混合式网络的一个研究重点. 发布/订阅模型实现了消息发布者和消息订阅者之间解耦的消息传递模式, 适用于混合网络之间的消息数据传输. 通过将发布/订阅模型应用于消息数据交换, 规范了消息数据的格式, 实现了对各类通信设备的灵活管理以及基于消息内容的动态数据路由; 并利用一种基于循环调度的动态负载均衡算法, 对低速率网络和高速率网络之间的性能进行合理调度, 提高了低速率网络的性能. 模拟实验结果表明, 在混合式网络中发布/订阅模型能实现可靠的消息数据交换, 在负载平衡算法下性能更好.     

基于KVM虚拟桌面的透明消息通道设计

云管理平台和虚拟机终端用户间的通信一般采用代理软件或插件,便捷性和抗干扰能力较低。针对该问题,利用基于内核的虚拟机(KVM)虚拟桌面,提出一种云服务提供节点(虚拟机所在物理主机)和虚拟机终端用户之间双向交互的透明消息通道设计方案。在云管理平台中建立消息控制端,用于接收和处理服务节点发往虚拟机终端用户的消息,并将消息转换为图像,使图像内容以位图像素数据格式的方式读出到特定文件中,作为消息发送模块的消息来源。通过修改KVM虚拟化平台中Qemu-KVM集成的VNC Server端源码,在源码中添加消息发送模块和反馈接收模块,将消息集成融入到虚拟机桌面图像中,并对VNC Client远程终端反馈的消息进行处理,从而得到一条对虚拟机自身系统透明的,可在云平台和终端用户间双向交互的消息通道。实验结果验证了该设计方案的可行性。