面向软件定义网络的服务功能链优化部署算法研究

针对网络功能虚拟化(NFV)环境下,现有服务功能链部署方法无法在优化映射代价的同时保证服务路径时延的问题,该文提出一种基于IQGA-Viterbi学习算法的服务功能链优化部署方法。在隐马尔可夫模型参数训练过程中,针对传统Baum-Welch算法训练网络参数容易陷入局部最优的缺陷,改进量子遗传算法对模型参数进行训练优化,在每一迭代周期内通过等比例复制适应度最佳种群的方式,保持可行解多样性和扩大空间搜索范围,进一步提高模型参数的精确度。在隐马尔科夫链求解过程中,针对隐含序列无法直接观测这一难点,利用Viterbi算法能精确求解隐含序列的优势,解决有向图网络中服务路径的优化选择问题。仿真实验结果表明,与其它部署算法相比,所提IQGA-Viterbi学习算法能有效降低网络时延和映射代价的同时,提高了网络服务的请求接受率。

     

基于SDN和NFV的业务链管理系统研究

首先针对目前业务链系统拓扑固化、扩展难的问题,分析业务链系统SDN化以及NFV化的需求。SDN的基本特征是转发与控制分离,集中的软件控制以及开放的编程接口,通过使用SDN技术,可以使得业务链系统具有灵活、拓扑可视、扩展方便等特性。然后,分析基于SDN技术实现业务链管理系统的技术架构,基于SDN的业务链系统主要由业务链管理系统、策略管理系统、SDN控制器、流分类器、SDN交换机等主要组件组成。文章对其中的关键技术进行了研究分析,最后,对业务链应用场景进行了研究,目前业务链系统在数据中心、Gi-LAN、接入网等场景有大量的应用需求。     

基于自动机的TCP流识别算法

为提升网络流识别性能,本文提出了一种TCP流识别算法.该算法基于传输控制协议(Transmission Control Protocol,TCP)下网络通信双方的交互过程构建双向流自动机,由该自动机根据TCP协议规则和网络流当前状态判断TCP流终止,同时以基于规则的过滤机制和超时策略为辅助措施,快速识别单包流和异常中断流.该算法内存开销、计算和内存总开销均低于经典算法固定超时策略(Fixed Timeout strategy,FT)和同类代表性算法两层自适应超时策略(Two-level Self-Adaptive Timeout,TSAT),同时该算法精度高于TSAT,且仅比默认精度标准略有下降.该算法基于协议规则识别TCP流,既保证了流的准确性,又节省了流的超时等待时间,而且算法尤其适合中流、小流和不规则TCP流比重较大的情况,使得识别系统在面临DDoS攻击、蠕虫爆发等网络异常时仍能正常运行.     

一种基于智能有限自动机的正则表达式匹配算法

本文提出了一种基于智能有限自动机(Smart Finite Automaton,SFA)的正则表达式匹配算法,在XFA的分支迁移边上增加额外的判断操作指令,消除XFA的回退迁移边,避免不必要的状态迁移操作.实验结果表明,SFA提高了正则表达式匹配的时空效率,与XFA相比,在存储空间开销上减少了44.1％,在存储器访问次数上减少了69.1％.     

Service Composition for Mobile Environments

Service Composition, that is, the development of customized services by discovering, integrating and executing existing services has received a lot of attention in the last couple of years with respect to wired-infrastructure or Internet web services. With the advancement in the wireless technology and rapid deployment of mobile devices, we envision that in the near future wirelessly connected mobile devices in a given vicinity will also provide services that can be leveraged in the composition process. This is particularly true of what have been described as “pervasive computing” environments. However, wired-infrastructure based service composition architectures are not designed to consider the various factors like mobility, device heterogeneity, resource variability and reliability in a mobile environment. In this paper, we describe the issues related to service composition in mobile environments and evaluate criteria for judging protocols that enable such composition. We present a distributed architecture and associated protocols for service composition in mobile environments that take into consideration mobility, dynamic changing service topology and device resources. The composition protocols are based on distributed brokerage mechanisms and utilize a distributed service discovery process over ad-hoc network connectivity. We present simulation results of our protocols, and compare them with a centralized service composition protocol traditionally used for wired-infrastructure environments. The results show that our approach clearly outperforms the existing centralized approaches, and that our protocols are able to adapt and better utilize the changing service topology and resources in a mobile environment.This work is supported in part by NSF awards 9875433 and 0070802, DARPA DAML program and IBM. Dipanjan Chakraborty is a Ph.D candidate and a research member of ebiquity research group at University of Maryland, Baltimore County (UMBC). His reserach is in the areas of mobile and pervasive computing environments, mobile and e-commerce, peer-to-peer systems with special interests in the fields of service discovery, information aggregation and composition, ad-hoc network application-centric routing, agent-based systems. He specializes in the development and modeling of distributed architectures to enable mobile and pervasive commerce in ubiquitous environments. His thesis is in the area of service discovery and composition for pervasive environments. He has been a fellow of IBM during the 3 years of his Ph.D candidacy. Anupam Joshi is an Associate Professor of Computer Science and Electrical Engineering at UMBC. Earlier, he was an Assistant Professor in the CECS department at the University of Missouri, Columbia. He obtained a B. Tech degree in Electrical Engineering from IIT Delhi in 1989, and a Masters and Ph.D. in Computer Science from Purdue University in 1991 and 1993 respectively. His research interests are in the broad area of networked computing and intelligent systems. His primary focus has been on data management for mobile computing systems in general, and most recently on data management and security in pervasive computing and sensor environments. He has created agent based middleware to support discovery, composition, and secure access of services/data over both infrastructure based (e.g. 802.11, cellular) and ad-hoc wireless networks (e.g. Bluetooth). He is also interested in Semantic Web and Data/Web Mining, where he has worked on personalizing the web space using a combination of agents and soft computing. His other interests include networked HPCC. He has published over 50 technical papers, and has obtained research support from NSF, NASA, DARPA, DoD, IBM, AetherSystens, HP, AT&T and Intel. He has presented tutorials in conferences, served as guest editor for special issues for IEEE Personal Comm., Comm. ACM etc., and served as an Associate Editor of IEEE Transactions of Fuzzy Systems from 99-03. At UMBC, Joshi teaches courses in Operating Systems, Mobile Computing, Networking, and Web Mining. He is a member of IEEE, IEEE-CS, and ACM. Tim Finin is a Professor in the Department of Computer Science and Electrical Engineering at the University of Maryland Baltimore County (UMBC). He has over 30 years of experience in the applications of Artificial Intelligence to problems in information systems, intelligent interfaces and robotics and is currently working on the theory and applications of intelligent software agents, the semantic web, and mobile computing. He holds degrees from MIT and the University of Illinois. Prior to joining the UMBC, he held positions at Unisys, the University of Pennsylvania, and the MIT AI Laboratory. Finin is the author of over 180 refereed publications and has received research grants and contracts from a variety of sources. He has been the past program chair or general chair of several major conferences. He is a former AAAI councilor and is a member of the the board of directors of the Computing Research Association. Yelena Yesha received the B.Sc. degree in Computer Science from York University, Toronto, Canada in 1984, and the M.Sc. and Ph.D degrees in Computer and Information Science from The Ohio State University in 1986 and 1989, respectively. Since 1989 she has been with the Department of Computer Science and Electrical Engineering at the University of Maryland Baltimore County, where she is presently a Verizon Professor. In addition, from December, 1994 through August, 1999 Dr. Yesha served as the Director of the Center of Excellence in Space Data and Information Sciences at NASA. Her research interests are in the areas of distributed databases, distributed systems, mobile computing, digital libraries, electronic commerce, and trusted information systems. She published 8 books and over 100 refereed articles in these areas. Dr. Yesha was a program chair and general co-chair of the ACM International Conference on Information and Knowledge Management and a member of the program committees of many prestigious conferences. She is a member of the editorial board of the Very Large Databases Journal, and the IEEE Transaction on Knowledge and Data Engineering, and is editor-in-chief of the International Journal of Digital Libraries. During 1994, Dr. Yesha was the Director of the Center for Applied Information Technology at the National Institute of Standards and Technology. Dr. Yesha is a senior member of IEEE, and a member of the ACM.This revised version was published online in August 2005 with a corrected cover date.     

基于非确定有穷自动机的网络入侵容忍系统研究

入侵容忍技术是继防火墙、存取控制、入侵检测技术之后的网络信息安全技术。当一个网络系统遭受非法入侵后，入侵容忍系统仍能厦时自我诊断、恢复和重构，并能为合法用户提供所需的全部或者降级的服务。本文依据非确定性有穷自动机理论，模拟计算机受到攻击时入侵容忍系统的工作过程，为入侵容忍技术的发展提供了一种方法。     

Chaining in a Loop System

This paper offers an approximate analysis of the chaining operation in a loop system. The system consists of a central processor andNterminals linked by a common communication channel. Chaining refers to the service protocol whereby the channel is assigned in sequence to each of the terminals. Messages from terminals to the central processor are sent as chains of segments. The object of the analysis is to display the relationships between the arrival processes, the channel allocation, and the response of the system.     

QoS-Aware Service Selection Algorithms for Pervasive Service Composition in Mobile Wireless Environments

The successful application of pervasive services running in mobile wireless networks and devices relies on its ability to provide efficient and cost-effective QoS (Quality of Service) support. This paper proposes a comprehensive QoS model specifically for pervasive services. It considers not only user-perceived factors but also mobile wireless network characteristics. The corresponding formula to calculate each QoS criterion is also devised. In particular, this paper formulates the QoS-aware service selection problem for pervasive service composition and proposes some solutions to the problem, i.e., global-search-based LOSSA (local optimal service selection algorithm) and limited broadcast based LOSSA-k. The evaluation results of the algorithms have shown the effectiveness of the QoS model and the efficiency of the proposed LOSSAs.     

基于QoS的语义Web服务组合优化算法

在面向服务的环境下,存在众多可满足服务请求者要求的服务组合方案,如何从中选出最优或近似最优方案是Web服务组合优化要解决的问题.设计出一个不必探索整个解空间的优化算法——SBG算法,为服务请求者提供最优选择方案.实验结果表明,这种方法可有效提高选择方案的求解效率.     

基于标注依赖图的服务组合方法

针对服务组合问题,提出了一种既考虑服务输入/输出又考虑服务前提条件/执行效果的结构化组合方法.根据服务请求查找服务组合路径,得到服务依赖图,然后通过依赖图中的依赖关系,把依赖图转化为结构化流程模型,除了标注结构化过程中的某些分支类型时,需要一些额外的用户干预外,整个过程是完全自动的.     

基于链接与计数的分组密码认证模式

为了提高分组密码算法的工作效率,提出了一种基于链接与计数(chaining & counter)的快速认证模式(CCTR),并论证了其安全性.速度测试表明CCTR模式比常用认证模式(例如CBC-MAC)快30%左右.依赖性测试表明:CCTR模式是完备的;其雪崩效应度约为0.999 3;其严格雪崩准则度约为0.992.频率测试表明CCTR模式的输出服从均匀分布.二进制矩阵秩、动向(run)、频谱、非重叠字(non-overlapping template)匹配、重叠字匹配、Maurer通用统计、Lempel-Ziv压缩、线性复杂度、系列、近似熵、累积和、随机游程以及随机游程变量等测试结果都满足要求.由此说明,CCTR模式具有很好的伪随机性,这就从统计评估的角度验证了CCTR模式的安全性.     

Action Patterns Probing for Dynamic Service Composition in Home Network

The concept of a smart home has been discussed in recent years. The major purpose is to make life more convenient, safe, and fun in various areas, including home automation, security, entertainment, and so on. In order to automate the interactions between the home users and devices or even between devices, the prediction of the home user’s actions and the integration of devices are very important. The UPnP Device Architecture defines the protocols for communication between the UPnP control point and devices. Exploiting UPnP techniques, home users can easily control intelligent devices through the control point. However, UPnP devices lack a composition mechanism to complete a novel application or value-added service.In this paper, an action patterns probing algorithm is proposed. We propose a dynamic service composition system which coordinates the primitive UPnP services at home. We can predict the action and the data flow with satisfactory accuracy. At first, we define data type ontology for UPnP devices to describe their service interfaces. Afterwards, the interface matching mechanism is employed to construct a service graph that describes which services can be composed together. And we have to analyze the record of user’s actions by using the service graph. Finally, we can find the devices which can be composed and worked together in common use. These devices can be composed dynamically by user’s habits and can be automated by our mechanism.     

基于Sub Web Service服务组合方法

在当前Web服务动态组合研究的基础上,提出了一种基于Sub Web Service的混合Web服务组合方法。结合静态与动态服务组合方法,将动态生成的服务组合计划描述保存在缓冲池中,当用户使用该组合服务时,系统将会在缓冲池中搜寻此服务组合,并为请求调用它。在组合算法中提出了基于Sub Web Service的组合算法,将多输入多输出的Web服务分解为多输入单输出的Sub Web服务执行服务组合。在避免重复动态组合服务的基础上,减轻了Web服务在组合过程中对多输出接口的依赖性,提高了系统服务效率。     

多层次组合Web服务筛选方法

为了能够更好地满足用户对服务组合的需要,文中首先给出了多层次纽合Web服务的筛选模型,并将多层次组合Web服务筛选模型分为五个步骤。在此基础上将Web服务的属性分为功能属性和非功能属性（即QoS属性）。通过划分服务类对功能属性进行描述,并按照服务类以QoS属性的类型进行服务组合。最后给出了多目标方案筛选方法及其实验。     

基于上下文的Web服务动态组合

合成现有Web服务以生成功能更强更复杂的组合服务成为必然,组合服务的上下文信息具有十分重要的作用.针对Web服务描述语言仅支持服务功能性描述的不足,扩展了非功能属性描述,定义了一个Web服务的上下文模型,设计实现了感知上下文和执行动态服务替换的组合服务平台,通过运行旅游服务对该平台系统进行了有效验证.     

基于MapReduce的并行Web服务自动组合

如何在大规模的Web服务集合中进行快速、高效的自动组合是当前Web服务组合研究与应用的难点.传统的Web服务自动组合方法大多建立在单机计算基础上,服务数量一旦过多,规划或搜索空间随之膨胀,组合效率低下.本文提出了一种分步分治、深度优先搜索的Top-k Qos服务组合算法,并采用MapReduce实现了分布式、并行的服务自动组合过程.实验结果表明,该方法在应对大规模的服务集合时,能快速、高效的提供满足用户需求的组合服务.     

基于QoS服务组合模型及应用研究

目前电信网络服务尚无服务质量控制机制,无法为用户提供区分服务和性能保证,使得特殊用户的网络服务质量受到严重影响。这就要求网络提供商对其服务组合的服务质量进行建模评估。文中主要讨论面向服务质量的服务组合方法和保证组合服务执行过程中业务功能和服务质量的方法。研究影响网络服务质量的相关参数。从用户的角度出发,建立服务组合的服务质量模型,并从用户和网络提供商两方面考虑,讨论改善服务质量的实际方法。     

复杂上下文感知的组合服务选择框架

泛在网络环境中,组合服务选择不仅需要考虑服务上下文还需要考虑用户上下文,而描述上下文属性的数据类型可能有实数、区间数、三角模糊数和直觉模糊集。为解决这种基于复杂上下文的组合服务选择难题,本文提出一个复杂上下文感知的组合服务选择框架,能够感知服务上下文和用户上下文。该框架基于智能主体设计,其中用户端框架完成用户上下文管理,服务端框架完成服务上下文管理和组合服务选择。通过相关工作的比较,验证了该框架的优越性和有效性。     

基于BPEL的Web服务组合仿真

介绍了Web服务组合实际应用中存在的问题,提出了基于工作流引擎的Web服务组合仿真思想.在分析BPEL的Web服务组合机制的基础上,利用开源的Active BPEL Designer工具建立BPEL流程, 实现了对智能卡注册过程业务中服务组合的仿真.