（p，k）约束流的可扩展性和QoS保证

在弱硬实时应用中,（p,k）约束可以提供更精细的QoS保证。为了提高（p,k）约束流的可扩展性,本文通过分析（p,k）约束,给出其分类依据,提出了相应的静态和动 态分类选择算法。实验结果表明,分类选择算法能够在性能和开销之间提供适当的折衷。     

多媒体网络中基于优先级调度的实时QoS研究

使用到达函数和服务函数描述不同优先级实时分组集的到达和处理,基于期望可调度性来刻画不同紧急程度的实时应用对延迟的要求。提出的方法能够实现实时QoS控制下的瞬时特性研究,判定某一时刻不同优先级分组集的可调度性。最后对指数分布服务时间下的不同优先级分组集的期望可调度性进行了研究,基于理论结论的数值结果和基于模型的模拟结果是一致的。     

实时控制系统中固定优先级调度的延迟与抖动控制

在基于固定优先级调度实时控制系统中,任务的延迟与抖动是影响系统稳定性的重要因素,提出一种基于可抢占时间阈值的延迟与抖动控制策略,给出一种保证系统可调度的最优闽值分配算法,并通过对任务延迟和抖动的分析量化出阀值分配后的争最大可能IO延迟及抖动.最后通过仿真实验验证了该策略的有效性.     

多优先级业务流局部化多路径QoS路由算法

全局QoS路由需要获知精确的全局网络QoS状态。由于不可忽略的传播延时和处理开销，节点所获得的QoS信息总是不精确的。研究了节点只了解部分的网络状态信息的情况下，当网络中存在具有优先级差异的多类别业务流时可行的局部QoS路由策略。仿真结果验证了算法的有效性。     

多优先级业务流的局部化多路径QoS路由算法

提出节点只了解部分网络状态信息的情况下,当网络中存在具有优先级差异的多类别业务流时可行的局部QoS路由策略,仿真结果验证了算法的有效性。     

多优先级业务流局部化多路径QoS路由算法

全局QoS路由需要获知精确的全局网络QoS状态。由于不可忽略的传播延时和处理开销，节点所获得的QoS信息总是不精确的。研究了节点只了解部分的网络状态信息的情况下，当网络中存在具有优先级差异的多类别业务流时可行的局部QoS路由策略。仿真结果验证了算法的有效性。     

一种基于标准化QoS公平的反馈控制实时调度算法

任务间的QoS公平性是多QoS级别的可调节动态实时调度系统中的一个重要问题.本文引入了标准化QoS公平的概念,在公平性中加入了任务的重要性因素,并构造了具有PID反馈控制环节的实时调度算法NF-QoS来达到标准化QoS公平.利用NF-QoS对系统截止期错失率进行实时采样,定期反馈给PID控制器,根据PID控制器计算的结果对各任务QoS级别进行调整,以保证各任务公平地得到处理器资源.实验结果表明,NF-QoS不仅合理地协调了系统利用率和截止期错失率,并且有效地提高了系统的QoS公平性.     

OFDMA系统中基于多业务QoS保证的机会调度算法

针对正交频分多址接入（OFDMA）系统中多业务资源调度问题,提出了一种基于服务质量（QoS）的最小性能保证的资源机会调度算法。以系统总吞吐量优化为目标,根据业务的QoS参量设计子载波调度参数,并利用图论中最大加权双向匹配问题的求解算法,在各业务间公平分配子载波资源,保证了各业务的QoS要求。仿真结果表明,所提算法在保证各业务的最低速率要求的前提下有效地提高了系统总吞吐量性能。     

802.16 Mesh模式下提供QoS保障的调度算法

无线Mesh网络是一种多跳分布式自组织网络。提出了一种IEEE 802.16 Mesh模式下的分布式QoS保障调度算法。该算法充分利用方向性天线带来的空分复用增益,并采用DiffServ机制保证QoS需求。仿真结果表明,所提算法不仅提高吞吐量,而且能够为不同类型的服务提供QoS保障。     

内容中心网络中基于内容感知的QoS保证机理探析

针对目前内容中心网络在QoS保证方面研究不足的问题,提出一种基于内容感知传输机制的内容优先级调度算法。该算法利用内容热度值判决内容数据包的优先级概率决定不同内容的调度顺序,能够根据网络环境动态调整传输队列。仿真结果表明,该机制在高优先级内容流量比例为20%的情况下,平均吞吐量相比于NDN常规传输机制性能提高约35%。     

QoS Management Through Adaptive Reservations

Reservation based (RB) scheduling is a class of scheduling algorithms that is well-suited for a large class of soft real-time applications. They are based on a bandwidth abstraction, meaning that a task is given the illusion of executing on a dedicated slower processor. In this context, a crucial design issue is deciding the bandwidth that each task should receive. The point we advocate is that, in presence of large fluctuations on the computation requirements of the tasks, it can be a beneficial choice to dynamically adapt the bandwidth based on QoS measurements and on the subsequent application of feedback control (adaptive reservations).In this paper, we present two novel contributions to this research area. First, we propose three new control algorithms inspired to the ideas of stochastic control. Second, we present a flexible and modular software architecture for adaptive reservations. An important feature of this architecture is that it is realised by means of a minimally invasive set of modifications to the Linux kernel.This work has been partially supported by the European OCERA IST-2001-35102 and RECSYS IST-2001-32515 projects.Luca Abeni is a Ph.D. in Computer Engineering at the Scuola Superiore SantAnna of Pisa (Italy). He graduated in Computer Engineering at the University of Pisa in 1998, and received a Ph.D. in Computer Engineering at the Scuola SuperioreS. Anna of Pisa in 2002. During 2000 he was a visiting student at the Carnegie Mellon University (Pittsburgh, PA), working with Prof. Ragunathan Rajkumar on resource reservation algorithms for real-time kernels. During 2001 he was a visiting student a Oregon Graduate Institute (Portland, OR) working with Prof. Jonathan Walpole on the support for time-sensitive applications in the Linux kernel. He has been working in Broadsat S.R.L. since 2003, developing audio/video streaming solutions and IPTV applications.Tommaso Cucinotta got the degree in Computer Engineering at the University of Pisa (Italy) in 2000. He got the Ph.D. inComputer Engineering at the Scuola Superiore SantAnna(SSSA) of Pisa in 2004 with a thesis titled Issues in authentication by means of smart-card devices. He held a course on cryptography in the International Master on Software Engineering organized at SSSA in 2002 and 2004. At the moment, he cooperates in research activities at the Scuola Superiore S. Anna in the areas of computer security and Quality of Service control for soft real-time systems.Giuseppe Lipari graduated in Computer Engineering at the University of Pisa in 1996, and received the Ph.D. degree in Computer Engineering from Scuola Superiore SantAnna in 2000. During 1999, he was avisiting student at University of North Carolina at Chapel Hill, collaborating with professor S.K. Baruah and professor K. Jeffay on real-time scheduling. Currently, he is assistant professor of Operating Systems with Scuola Superiore SantAnna. His main research activities are in real-time scheduling theory and its application to real-time operating systems, soft real-time systems for multimedia applications and component-based real-time systems.Luca Marzario is a Ph.D. student in Computer Engineering at the Scuola Superiore S. Anna of Pisa, Italy. In 2002, he graduated in Computer Engineering at University of Pisa. His main research interest include real-time systems scheduling, aperiodic service mechanism, feedback-scheduling, QoS in multimedia systems, Linux kernel and Real Time Linux executives (RTAI, RTLinux).Luigi Palopoli graduated in Control Engineering at the University of Pisa in 1998 and received his Ph.D. degree in Computer Engineering in 2002 from the Scuola Superiore S. Anna, Pisa. During 2001 he was a visiting scholar at the Department of EECS, University of California at Berkeley, where he worked on design techniques for real-time controllers under the supervision of Professor Alberto Sangiovanni-Vincentelli. He is currently Assistant Professor at the Scuola Superiore S. Anna (Pisa). His main research activities include Quality of Service control, control of systems under communication and computation constraints and design of embedded systems.     

基于简单反馈的混合静态/动态节能弱硬实时调度算法

随着能耗问题目益显著,节能实时调度成为实时调度领域研究的热点.由于混合静态/动态节能弱硬实时调度算法基于最坏情况执行时间计算任务的执行速度,因此限制了节能效果,文中针对这一问题,提出一种新算法,通过引入简单反馈机制,估计任务的实际执行时间,通过任务划分,降低任务的整体执行速度,延长执行时间,进而达到高效节能的目的.实验表明,当平均情况执行时间低于最坏情况执行时间较多时,新算法优于原始算法,最多可节能60%～70%,最少可节能约10%.算法的不足之处在于当平均情况执行时间接近最坏情况执行时间时,新算法比原算法更耗能.     

基于云模型的网格QoS参数归约与调度优化

针对网格资源调度中用户对QoS的定性描述,利用云模型实现资源调度中的QoS匹配。深入分析了QoS参数云的特征,提出了QoS云处理模型,通过该模型,将离散的多个QoS参数归约到一个定性的概念上;设计了实现参数归约的体系结构;给出了基于定性概念的资源调度算法。实验表明,所提出方法在资源调度率和吞吐量以及系统资源的利用效率等方面体现出良好的特性,实现了基于定性概念的调度,达到了优化调度的目的。     

嵌入式操作系统的硬实时微内核设计

目前的嵌入式实时操作系统存在着内核结构较为复杂、臃肿、稳定性不高、对硬实时应用支持不足等问题.针对这些问题,结合现有的操作系统内核理论及嵌入式实时系统的特殊需求,通过组件化的设计,将内核划分为核心态基本模块及用户态扩展模块,提供多种进程间通讯(IPC)方式,并引入独特的硬实时抢占式调度机制,设计出一种精炼、稳定的硬实时微内核.通过这种设计思路成功开发出了昊鹏(Hopen)操作系统新版内核,应用于最新的3G手机产品上,取得了非常好的效果.     

面向QoS保障的虚拟可信服务*

针对服务计算模型中对应用的可信特性支持不足问题,提出面向QoS保障的虚拟可信服务VTS,并依据VTS设计了一种基于反馈控制的自适应QoS保障机制。首先将VTS的QoS保障转换为反馈控制问题,给出一种自适应QoS保障框架,再在此框架下对VTS的QoS维护过程和策略进行建模,设计和实现了相应的动态组建、调节算法和实时评估策略。最后通过仿真实验的结果分析表明,该保障机制能够有效地增强VTS对服务实体运行时的QoS保障能力。