网格信息服务互操作策略研究

针对当前网格中间件中信息服务系统的差异性和不可互操作性,提出了一种基于插件和虚拟层的网格信息服务互操作策略。该策略可以屏蔽底层网格中间件信息服务部件的异构性,为终端用户和作业调度模块提供统一的服务访问接口。通过在CSGrid网格互操作项目中的成功应用,该策略体现了以下特点:可在互操作的同时,保证各网格系统的独立性;只需开发一些插件就可扩展到更多的网格平台,具有良好的可扩展性;能够提供一定的可靠性保证和服务质量保证。     

基于策略的可控服务发现与动态路由模型

如何改进现有服务发现模型使之适应动态可变的服务运行环境并选择最符合用户需求的Web服务正在引起研究领域关注.提出了一种基于策略的可控服务发现与动态路由模型(P-WSDRM).该模型支持抽象服务、服务实例和服务发现者的属性定义,支持携带属性描述信息的服务发布与发现,引入了策略判定机制,支持服务发现者基于已定义的策略进行服务发现和实例路由.目前已经于Linux平台和目录服务实现了该模型的一个原型系统.     

一种基于服务自适应的系统可生存性策略

网络系统的可生存性已经成为一个热门的研究主题,该研究课题的思想是要保障系统关键服务的有效性.提出一种服务自适应策略.自适应策略受生物免疫原理的启发,通过控制网络服务的复制、迁移和消亡来提高系统的可生存性.模拟结果显示网络服务在动态负载和发生故障的情况下具有扩展性和鲁棒性,反映了系统具有一定的适应能力和生存能力,说明该策略具有可行性和有效性.     

支持成本效益分析的Web服务容错策略规划方法

基于各种冗余策略的Web服务容错方法被广泛应用于面向服务的关键性系统的可靠性和性能保障。已有的Web服务容错策略选择方法只考虑服务响应时间和失效率等与服务性能相关的因素,而忽略了服务调用和补偿费用等成本问题,导致所选择的策略可能在成本效益方面存在不足。针对这一问题,提出了一种支持成本效益分析的Web服务容错策略规划方法。该方法针对九种常用的Web服务冗余策略给出了综合考虑调用和补偿费用的成本预测公式,并在此基础上给出了支持成本效益分析的Web服务容错策略选取算法。针对该方法的验证实验表明,所提方法能够显著提高Web服务容错策略的有效性,并且所产生的额外时间开销在可接受范围内。     

OSGi平台服务动态更新的发布/订阅策略

针对7×24小时运行的系统,不能停止系统运行进行更新或维护的问题,根据OSGi( Open Service Gateway Initiative)框架的特点,在分析已有的服务动态更新方法基础上,克服已有策略不能保证服务更新前后的相互一致性和服务状态的一致性,以及更新策略实现过程中的编程模型不佳问题,提出基于发布/订阅模型的服务动态更新策略.以静止状态理论为依据,利用Petri网形式化描述服务动态更新过程,从理论上验证该策略的有效性;同时,以Event Admin Service作为策略的实现工具,具有良好的编程模型,从而实现OSGi平台上的动态演化,实验结果表明该策略具有可行性.     

调度服务在虚拟试验系统中的研究与应用

为了提高虚拟试验系统的执行效率,保证系统负载平衡,提高系统的实时性和可靠性,研究并设计了一种实时任务调度服务;该调度服务以节点任务调度算法和迁移策略为基础,提供了一种动态的、可靠的任务分配策略和负载均衡机制,使虚拟试验的任务得到了合理分配和调度;以节点容纳度作为试验参数进行多次试验,试验结果表明:该调度服务在虚拟试验系统中能够动态维持各个节点的负载平衡,降低了任务平均响应时间,提高系统的实时性和可靠性,达到了提高虚拟试验系统整体性能的目的。     

一种简化的基于First-Fit的Backfilling调度策略——RB-FIFT

机群作业管理系统是机群系统的重要组成部分,而作业调度策略又是机群作业管理系统的核心内容。作业调度策略的选择不仅关系到机群系统的利用率,还决定用户作业的响应速度和平均执行时间。在节点分配基于空间共享策略(Space-Sharing)的机群系统中,传统的基于先来先服务的First-Fit调度策略虽然能够提高机群系统的利用率,却容易引起饥饿问题。文章基于传统的先来先服务的调度策略,提出了一种简化的Backfilling算法,简要叙述了该算法的设计和实现,最后根据模拟实验结果,从用户和系统的角度分析了该算法的性能。     

一种基于策略的网络管理体系结构

提出了一个采用CORBA实现的基于策略的网管系统的体系结构,它利用了CORBA易于集成、可扩展性好,互操作性强等特点,结合CORBA所提供的系统服务研究了分布异构网络环境下该体系结构的策略描述、策略决策和策略执行的实现技术。该体系结构还能很好地实现对现有网络的策略管理。     

P2P流媒体直播系统中的合作节点选择策略

合作节点选择是P2P流媒体直播系统中的核心问题之一。针对此问题进行深入分析,提出了一种基于服务能力的启发式合作节点选择策略,根据节点的服务能力对其在系统中的位置进行自适应调整,使得节点最终形成层状结构,且服务能力越高的节点位于越高层。仿真结果表明,与随机选择策略相比,该策略能够显著降低系统传输延迟。     

切换服务系统的稳定性及交叉口信号配时

当缓冲器内工作量较大时,服务器的“固定顺序清空切换”服务策略使得清空缓冲器内工作量需要较长时间,从而其他缓冲器未能及时得到服务.为此本文提出服务器“带服务时间上限的固定顺序切换”服务策略,服务器在该服务策略下分配给每个缓冲器服务时间上限,以保证各缓冲器公平地得到服务.在满足缓冲器总负荷小于各缓冲器服务时间调整因子的最小值与最大值之比的条件下,证明了所提出服务器服务策略下的切换服务系统是周期稳定的.进一步将交叉口信号配时建模为切换服务系统,并应用所提出的服务器服务策略研究了交叉口信号配时策略.通过对交叉口实际数据的仿真验证了本文方法的有效性.     

基于软件定义网络的媒体分发网络的接入控制

针对软件定义网络（SDN）环境下的媒体分发网络的接入控制问题,提出了一种综合考虑服务节点和传输链路服务性能优化问题的接入控制方案。该方案利用SDN控制器对路由器的直接管控和对全网的感知能力,对应用层上服务节点服务性能和网络层上服务节点到用户之间的传输链路的服务性能进行联合优化,减少了链路拥塞对数据传输和用户服务质量的影响。首先,为SDN服务系统的接入控制过程建立部分可观Markov决策过程（POMDP）模型;然后,使用基于观测的随机策略作为系统的接入控制策略;最后,通过策略梯度算法对接入控制策略进行优化,求解出模型的最优策略。仿真结果表明,与尽力而为的服务策略相比,基于POMDP模型的最优接入控制策略使系统性能提高了10%,验证了所提方法的有效性。     

Using aggregation to construct periodic policies for routing jobs to?parallel servers with deterministic service times

The problem of routing deterministic arriving jobs to parallel servers with deterministic service times, when the job arrival rate equals the total service capacity, requires finding a periodic routing policy. Because there exist no efficient exact procedures to minimize the long-run average waiting time of arriving jobs, heuristics to construct periodic policies have been proposed. This paper presents an aggregation approach that combines servers with the same service rate, constructs a policy for the aggregated system, and then disaggregates this policy into a feasible policy for the original system. Computational experiments show that using aggregation not only reduces average waiting time but also reduces computational effort.     

An application of modern control theory to a time-dependent queuing system for optimal operation

‘ Bang-bang ’ optimal closed-loop service policy for a time dependent M/M/l queuing system is derived using optimal control theory. The policy is based on probabilistic (and not stochastic) behaviour of the queue. Computational results are obtained for an illustrative example with non-bang-bang service policy using the conjugate gradient algorithm with bounded control variables. It is interesting to note that the optimal service policy is not sensitive to the arrival rate but to the mean service cost of a customer.     

Optimal control of the service rate of an exponential queuing network using Markov decision theory

An exponential queuing network of two service stations is considered. The service rate at each station can either be fast or slow. A policy in which the service rate to be used at each station is specified as a function of the state of the system is investigated in order to minimize a given objective cost function. Howard's policy space algorithm is used to characterize the form of the optimal policy. Results show that the optimal policy is a simply connected one and of montone structure.     

基于SOA的策略服务平台设计与实现

本文通过分析现有基于策略的网络管理系统架构,结合SOA思想设计并实现基于SOA的策略服务平台,并将该平台应用到基于策略的异构网络管理系统中,为其提供策略服务,实现了对异构网络资源的高效自治管理。     

Analysis of two-tier public service systems under a government subsidy policy

Some public service systems such as healthcare systems consist of both free public service provider with a long wait time and paid private service provider with a short wait time. Such service systems are often called a two-tier service system. In general, more customers will choose the free service provider (SP). To reduce the congestion in the free system, the government may encourage customers to use the pay system by offering them a subsidy. This paper studies whether such a subsidy can reduce the free system’s waiting time and improves the social welfare. While the objective of the free system is to maximize its own total customers’ utility, the objective of the pay system is to maximize its profit. We develop a mixed duopoly game to analyze the Nash equilibrium for the competition between the free and toll systems. Two scenarios with unregulated and regulated prices are considered. When the pay system price is unregulated (the private SP can set prices freely), we find that if customers are less sensitive to the service delay, the subsidy policy can effectively reduce the expected waiting time for the free system and increase the customer utility surplus of the entire two-tier system. However, if customers are more sensitive to the service delay, the subsidy policy may have the opposite effect. When the pay system price is regulated (the price determined by government), the subsidy policy can effectively reduce the expected waiting time for the free system and improve the social welfare of the two-tier system. And there exists an optimal regulated price to maximize the social welfare of the entire public service system.     

A methodological framework for formulating information policy

Effective operation, control and management of the information service within an organization are essential to its success. Nevertheless, many management decisions that effect this service are made by default, or through a play of forces, rather than in accordance with general management policy. This paper presents a framework for formulating an information policy in an organization. Its architecture is built up of four components: information policy goals, measures to determine how well these are achieved, policy actions and alternatives, and the information services system itself.A list of policy actions, outlining the decisions which should be made by management, in cooperation with information service department and its users, is organized into an Information Policy Matrix. This provides a structured approach.     

Optimal pricing policies for services with consideration of facility maintenance costs

For survival and success, pricing is an essential issue for service firms. This article deals with the pricing strategies for services with substantial facility maintenance costs. For this purpose, a mathematical framework that incorporates service demand and facility deterioration is proposed to address the problem. The facility and customers constitute a service system driven by Poisson arrivals and exponential service times. A service demand with increasing price elasticity and a facility lifetime with strictly increasing failure rate are also adopted in modelling. By examining the bidirectional relationship between customer demand and facility deterioration in the profit model, the pricing policies of the service are investigated. Then analytical conditions of customer demand and facility lifetime are derived to achieve a unique optimal pricing policy. The comparative statics properties of the optimal policy are also explored. Finally, numerical examples are presented to illustrate the effects of parameter variations on the optimal pricing policy.     

Modeling and analysis for streaming service systems

The proliferation of streaming service system in various application areas gains increasing importance and also poses more challenges in the research of streaming service system. In this paper, we propose a novel dynamic model composed of a set of differential equations to describe the evolution of streaming service systems. And in the model, we focus on how the policies for admission control and peer selection influence on the system. We first introduce a flexible abstraction of streaming service systems. The abstraction is generally enough to capture the essences of streaming service systems with different structures, physical characteristics, software protocols and client behaviors. Then, by analyzing the state which is defined as the number of requests, a novel dynamic model is developed in microscopic scale to characterize the behaviors of streaming service systems. The model proposed in this paper demonstrates the interactions between clients and servers and also between different servers. The interactions are primarily influenced by the admission control policy and peer selection policy. Finally, some experiments are designed to verify the validation and reasonability of the model.     

服务率不确定的单站点传送带给料加工站系统鲁棒优化控制

研究了服务率不确定情况下的单站点传送带给料加工站(CSPS)系统的鲁棒优化控制问题。在仅知服务率区间的条件下,以CSPS系统的前视距离作为控制变量,将鲁棒优化控制问题建模成不确定参数的半马尔可夫决策过程(SMDP)的极大极小优化问题,在状态相关的情况下,给出全局优化算法进行鲁棒控制策略求解。首先,运用遗传算法求解固定策略下的最差性能值;其次,根据求解得到的最差性能值,运用模拟退火算法求解最优鲁棒控制策略。仿真结果表明,服务率不确定的CSPS系统的最优鲁棒性能代价与服务率固定为区间中值系统的最优性能代价相差不大,并且随着不确定区间的缩小,两者的差值越小,说明了全局优化算法的有效性。