可用性感知的服务动态选择机制研究
发布时间：2018-08-27 13:43
【摘要】：随着Internet与Web技术的高速发展及广泛应用,如何从众多的服务功能相近或相同但QoS属性值不同的Web服务中选择出最能满足用户需求的Web服务,已经成为相关领域的一个研究热点。传统的Web服务选择技术虽然能够满足用户的-般功能性服务请求,但是由于已有的服务选择技术没有充分考虑服务组件的动态性以及服务请求高峰期产生的服务器运行负载不均衡等服务的非功能性属性导致服务不可用问题,因此不能满足用户有确定服务质量等级要求的服务请求。论文的核心工作就是围绕这些问题,使用负载感知每一服务请求方法以及可用性感知的服务依赖关系的建模等自适应技术,针对面向服务的应用系统设计以及保证服务选择质量的系统优化等方面作了较深入的研究。 本文的主要工作有以下4个方面： 1.提出可用性感知的服务选择模型;本文深入研究并分析了服务选择相关理论以及自适应技术。对Web服务及其服务等级协议进行建模,分析任务执行Web服务的工作流程、描述语言以及其QoS属性计算,设计了可用性感知的服务选择模型,使任务感知服务组件的可用性。 2.基于可用性感知服务选择模型给出可用性感知的全局最优的服务动态选择算法。包括以下3个方面的解决方案： (1)负载均衡策略,解决任务执行过程中某一服务载荷过重而引起的服务性能下降的问题。 (2)可用性记录评价服务方法,针对当前Web服务组件的动态性导致服务失败的问题,设计可用性服务组件模型,对服务组件每一次被调用执行的完成情况进行历史累加记录,并设计可用性记录调整算法,用可用性记录作为服务选择时一种选择约束条件。 (3)可用性感知的服务动态选择算法,基于负载均衡策略与可用性记录评价服务方法,提出了可用性感知的服务动态选择算法,利用0-1整数线性规划优化流程,并进行设计与实现,通过仿真实验证明其有效性。 3.设计实现了一个支持可用性感知的的服务动态选择应用原型,给出了该原型系统的体系结构,以及系统实例分析。 4.系统仿真实验与分析,对上述给出的3个算法进行了3组仿真实验,通过实验,验证了原型系统DSCS可以避免服务组件动态性对系统服务带来的不利影响,并能保证在访问量较大时系统具有负载均衡的能力,较好地保证了面向服务应用系统能够为用户提供连续一致的服务,即验证了本文方法的有效性。
[Abstract]:With the rapid development and wide application of Internet and Web technology, how to select the Web service which can best meet the needs of users from many Web services with similar or same service functions but different QoS attribute values has become a research hotspot in related fields. Although the traditional Web service selection technology can satisfy the user's functional service request, However, the non-functional properties of the service, such as the dynamic service component and the service load imbalance caused by the peak service request, have not been fully taken into account in the existing service selection technology. Therefore, it can not satisfy the service request that the user has defined the quality of service grade request. The core work of this paper is to use adaptive techniques such as load aware of each service request method and usability aware service dependency modeling around these problems. In this paper, the design of service oriented application system and the optimization of service selection quality are studied. The main work of this paper is as follows: 1. A usability aware service selection model is proposed, and the theory of service selection and adaptive technology are studied and analyzed in this paper. Modeling the Web service and its service level protocol, analyzing the workflow of the task executing Web service, describing language and its QoS attribute calculation, designing the service selection model of usability awareness. Make task aware of the availability of service components. 2. Based on the usability aware service selection model, a globally optimal dynamic service selection algorithm for usability awareness is presented. The solutions include the following three aspects: (1) load balancing strategy to solve the problem of service performance degradation caused by overloading of a service during task execution; (2) availability record evaluation service method. Aiming at the problem of service failure caused by the dynamic nature of the current Web service component, this paper designs a usability service component model, records the completion of each invocation of the service component, and designs an algorithm for adjusting the availability record. The availability record is used as a selection constraint in service selection. (3) A service dynamic selection algorithm based on usability awareness, which is based on load balancing strategy and availability record evaluation method. This paper presents a dynamic service selection algorithm based on usability perception. It optimizes the process by 0-1 integer linear programming, and designs and implements it. The simulation results show that the algorithm is effective. A service dynamic selection application prototype supporting usability awareness is designed and implemented. The architecture of the prototype system and the analysis of system examples are given. 4. System simulation experiment and analysis, three groups of simulation experiments on the above three algorithms are carried out. Through the experiments, it is verified that the prototype system DSCS can avoid the adverse impact of the dynamic service component on the system service. It can ensure that the system has the ability of load balancing when the traffic is large, and the service-oriented application system can provide continuous and consistent service to the user, which proves the effectiveness of the method in this paper.
【学位授予单位】：青岛大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.09
本文编号：2207468
[Abstract]:With the rapid development and wide application of Internet and Web technology, how to select the Web service which can best meet the needs of users from many Web services with similar or same service functions but different QoS attribute values has become a research hotspot in related fields. Although the traditional Web service selection technology can satisfy the user's functional service request, However, the non-functional properties of the service, such as the dynamic service component and the service load imbalance caused by the peak service request, have not been fully taken into account in the existing service selection technology. Therefore, it can not satisfy the service request that the user has defined the quality of service grade request. The core work of this paper is to use adaptive techniques such as load aware of each service request method and usability aware service dependency modeling around these problems. In this paper, the design of service oriented application system and the optimization of service selection quality are studied. The main work of this paper is as follows: 1. A usability aware service selection model is proposed, and the theory of service selection and adaptive technology are studied and analyzed in this paper. Modeling the Web service and its service level protocol, analyzing the workflow of the task executing Web service, describing language and its QoS attribute calculation, designing the service selection model of usability awareness. Make task aware of the availability of service components. 2. Based on the usability aware service selection model, a globally optimal dynamic service selection algorithm for usability awareness is presented. The solutions include the following three aspects: (1) load balancing strategy to solve the problem of service performance degradation caused by overloading of a service during task execution; (2) availability record evaluation service method. Aiming at the problem of service failure caused by the dynamic nature of the current Web service component, this paper designs a usability service component model, records the completion of each invocation of the service component, and designs an algorithm for adjusting the availability record. The availability record is used as a selection constraint in service selection. (3) A service dynamic selection algorithm based on usability awareness, which is based on load balancing strategy and availability record evaluation method. This paper presents a dynamic service selection algorithm based on usability perception. It optimizes the process by 0-1 integer linear programming, and designs and implements it. The simulation results show that the algorithm is effective. A service dynamic selection application prototype supporting usability awareness is designed and implemented. The architecture of the prototype system and the analysis of system examples are given. 4. System simulation experiment and analysis, three groups of simulation experiments on the above three algorithms are carried out. Through the experiments, it is verified that the prototype system DSCS can avoid the adverse impact of the dynamic service component on the system service. It can ensure that the system has the ability of load balancing when the traffic is large, and the service-oriented application system can provide continuous and consistent service to the user, which proves the effectiveness of the method in this paper.
【学位授予单位】：青岛大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.09
【参考文献】
相关期刊论文 前2条
1 蒋哲远;韩江洪;王钊;;动态的QoS感知Web服务选择和组合优化模型[J];计算机学报;2009年05期
2 周坤,邓保华,林齐圣,姚丹霖;面向Web Services动态复合的流程自动化系统的研究与实现[J];计算机应用;2005年01期
，本文编号：2207468
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2207468.html
