业务驱动的光子网格关键技术研究

通过对光子网格资源管理新机制、光子网格中间件和光子网格用户网络接口等关键技术的研究，建立业务驱动的光资源动态调度方法，给出大块数据传输及视频点播（VOD）服务在可控光路与基于OBS的网格（Grid over OBS）模式下的业务应用模式。理论与实验证明业务驱动的光子网格关键技术的实现，使得光子网格能很好地为密集数据业务服务。     

关于网格中间件的研究

网格中间件是网格关键技术之一。研究的目的是为用户提供具有同一编程接口的虚拟机器,支持复杂应用问题的求解和广域网上各类资源的共享。美国、英国都很重视这方面的研究,NMI中间件计划、Internet2中间件计划都取得了很大进展。设计网格中间件要具备资源动态监测、屏蔽节点异构、优化资源选择和协同计算等功能。网格中间件包括资源监测组件、计算服务组件、网格安全组件、容错服务组件、信息服务组件和应用调度组件。成功的网格中间件典型范例说明,它可以更好地支持系统为用户提供高品质的服务以满足需求。     

基于移动Agent的网格资源管理模型的研究

资源管理是网格中间件的重要组成部分，负责对网格资源进行组织、分配、调度，关系到资源分配、调度的效率，进而影响网格的整体性能。本文对当前关于网格资源管理的模型进行了评价，分析了移动代理用于网格计算的技术优势，提出了基于移动Agent的网格资源管理模型。     

CROWNBench:基于可定制合成负载的网格性能测试系统

介绍了一个利用可定制的合成负载对网格进行性能测试的系统-CROWNBench。CROWNBench所提供的机制允许测试者定制网格性能测试中所运行网格应用的结构以及负载合成方式,自动合成测试负载,并在测试开始后模拟真实网格作业提交方式向被测网格系统加载。CROWNBench依据本系统所建立的网格负载模型,定义了网格性能指标,并将自动测试后得到的性能指标测量结果作为衡量网格性能的依据。应用本系统对部署CROWN网格中间件的网格系统进行性能测试的初步实验表明,本系统达到了预期的设计目标,即以高效、灵活、可控、可复现、自动化的方式完成网格系统的性能测试。     

Web Service网格计算综合型应用框架的研究

随着网格技术和WebService的不断发展,网格技术与WebService越来越趋融合。针对WebService在网格计算任务调用中请求和分发两种应用框架的单向性,文章提出了一种新的综合型应用框架,该框架可以更有效地在工作代理和网格资源中分配工作任务。最后用实验证明了这三种框架的执行时间差异。     

网格计算中间件的研究

在整个网格体系结构中。中间件层是网格计算中的核心。因此网格计算中间件就成为了最关键的部分。本文阐述了构建网格所必需的网格计算中间件的功能和组件模型,并着重对当前较为流行的中间件GT4进行了研究和探讨。最后对网格计算中间件的发展进行了展望分析。     

网格计算中间件的研究

在整个网格体系结构中,中间件层是网格计算中的核心,因此网格计算中间件就成为了最关键的部分.本文阐述了构建网格所必需的网格计算中间件的功能和组件模型,并着重对当前较为流行的中间件GT4进行了研究和探讨,最后对网格计算中间件的发展进行了展望分析.     

嵌入式视讯终端中编解码中间件模型的研究

利用中间件技术原理,在底层硬件平台与上层应用软件之间设计一个编解码中间件,对底层硬件的操作进行封装、对上层应用软件提供标准的调用接口。当底层硬件发生改变时,只对编解码中间件进行简单的修改,中间件对上层提供的标准的接口不变,上层应用软件几乎不做任何改变,加快了视讯终端产品的开发速度。     

一种基于工作流的网格中间件体系结构

网格计算将是计算机技术未来发展的重点,而面向应用的网格中间件将是网格能否大规模应用的关键.提出一种基于工作流的网格中间件体系结构,利用工作流管理系统进行解析和控制流程,把所有的应用统一到网格服务的概念之下,建立一种统一的应用集成方案,以此来解决异构环境下的协同应用问题.     

基于ARM的嵌入式RFID中间件系统设计

无线射频识别(Radio Frequency Identifi cation,RFID)技术作为新一代识别技术的代表,近年来发展迅猛。随着RFID技术的深入发展和应用,RFID系统的中枢和核心组成部件的中间件成为了研究的热点。RFID中间件是连接底层设备和上层应用的桥梁,可以实现RFID读写器与企业应用的连接。嵌入式RFID中间件构架于嵌入式系统的弱计算环境中,它有别于运行在一般计算机上的软件中间件,而是在嵌入式系统上实现RFID中间件功能,使中间件可以用于各种系统集成。     

Application Deployment on Catallactic Grid Middleware

An architecture based on a decentralized market view integrates grid applications with catallactic middleware. A prototype application showed the concept's feasibility, as well as the middleware's effectiveness in balancing query-request workload across multiple grid services. Grid computing researchers have shown significant interest in using an economic paradigm for exchanging grid resources and services. With this approach, applications can use a market mechanism to schedule services access, thus giving them a fair, efficient way to share resources in high-demand periods. Most existing approaches rely on centralized brokers to coordinate resource access, and they're typically implemented over existing grid middleware. We propose an alternative approach based on F.A. Hayek's Catallaxy mechanism. Catallaxy's free-market, self-organizing coordination mechanisms adjust prices within the market based on the particular demands for a specific (scarce) resource     

Pivotal Technology Research of Grid Based on Mobile Agent

Grid Based on Mobile Agent is a new grid scheme. The purpose of the paper is to solve the pivotal technology of Grid Based on Mobile Agent ( GBMA) combined with thought of Virtual Organization ( VO). In GBMA, virtual organization is viewed as the basic management unit of the grid, and mobile agent is regarded as an important interactive means. Grid architecture, grid resource management and grid task management are the core technology problem of GBMA. The simulation results show that Inter- VO pattern has the obvious advantage because it can make full use of resources from other virtual organizations in GBMA environment.     

High Efficiency and Light Mobile Electronic Business System Based on Mobile Agent Middleware

With the development of wireless network, many people want to carry out commerce activity because of its convenience and mobility. It is necessary to build high efficiency and collaborative transaction system due to high delay and burst error in wireless network. Unfortunately, it is difficult for traditional technology to achieve high efficiency in low bandwidth network and new technology is expected. Large analysis of collaborative commerce system shows that there are some common design elem…     

基于移动代理和信任机制的网格作业管理

介绍了网络计算环境下的作业管理,引入了移动代理技术和信任机制,并进行了深入分析,对所用到的关键技术进行了详细的解释,说明了基于移动代理和信任机制的网格作业管理运行步骤.     

网格环境下数据访问与集成的研究与实现

OGSA-DAI WSRF致力于建造通过网格访问和集成来自不同的孤立数据源的中间件,符合基于WSRF的网格标准。介绍了OGSA-DAI WSRF2.2中间件的开发背景、体系结构、重要术语及关键技术,分析基于WSRF的数据访问和集成服务框架,给出基于OGSA-DAI WSRF的数据访问与集成方法及应用实例。     

A service-integrated sensor network middleware applied to industrial solutions of IoT related

With the deployment and application of Internet of Things, middleware for sensor network has become an increasingly important factor to facilitate the programmer task and bridge the gap between the applications and the underlying infrastructure. Most existing middleware for sensor networks lack attention for integrating services into a generic architecture and present inflexibility that make them useless in the context of industrial solutions of Internet of Things. To solve this problem, we proposed a new middleware deployment model and present layered and distributed architecture utilizing our “Ends to the Middle” approach. Actual application of middleware demonstrates that our proposed architecture is highly modular and efficient, offers good performance in complex application scenarios of Internet of Things.     

下一代网络软件技术的发展趋势

在下一代网络（NGN）的总体框架下，从普适计算、网格、中间件以及技术融合上阐述了软件技术的发展趋势．     

Distributed policy processing in active‐service based infrastructures

More and more applications in the Internet are requiring an intelligent service infrastructure to provide customized services. In this paper, we present an infrastructure, which can transparently and effectively provide customized active‐services to end users and dynamically adapt to changing customized policies in large distributed heterogeneous environments. The infrastructure consists of two components: the policy agent and middleware box. Particularly, our technologies include: (1) Generic active‐service based infrastructure, where the policy agent can integrate policies requested by applications, and middleware boxes can transparently execute services and (2) Distributed policy processing in the middleware box. We study two policy partitioning schemes to achieve conflict‐free policies for distributed policy processing and guarantee the correctness of the policy execution. We conduct extensive performance evaluations on different schemes proposed. Our experimental results demonstrate that our policy partitioning schemes can effectively generate partition‐capable and conflict‐free policy sets. The evaluation results also show that distributed policy processing can achieve over 70% increase in performance/price ratio with proper assignment of the policy distribution degree compared to a purely centralized approach. Copyright © 2005 John Wiley & Sons, Ltd.