面向集团企业的数据集成模型构建方法研究

发布时间：2018-01-09 01:15

本文关键词：面向集团企业的数据集成模型构建方法研究　出处：《天津大学》2013年博士论文　论文类型：学位论文

【摘要】：在我国当前的经济环境中，集团型企业具有举足轻重的作用。大型企业往往组织结构形态复杂、具有多级下属分支机构、人员众多而且隶属关系复杂、业务分布在多个行业、经营地域分布广、管理模式多样、需要全方位的协同工作等特点。集团企业的大量信息散落在不同的子公司、不同部门、人员等处，信息的来源复杂，，信息量非常大，整合性很差，为企业的生产管理的决策带来具大的挑战。随着云计算的落地生根和技术的发展成熟,企业管理型应用向云计算环境下移植和部署的大背景下，集团企业迫切需要通过多数据源以及多数据中心的协作，实现将现有不同信息系统中分布且异构的数据集成起来。数据集成能够简化企业的业务流程和工作成本，实现企业从数据中获取商业利益的目的。所以数据集成已成为企业的一项战略性工作，是提高企业核心竞争力的重要基石。通过数据集成可以使集团成员企业共享分布式数据，使新业务的开拓，集团领导的监管，风险的防范成为可能。本文提出了云计算下基于本体的异构数据集成模型，该模型基于本体理论，按照云环境特点构建，支持各种传统存储和云存储的数据集成，满足企业用户对高并发访问以及对海量数据高效存储的需求，同时还要满足集团企业对存储数据高扩展性和高可用性以及对数据事务一致性的需求。完成集团企业信息化应用在云上的部署和迁移，促进企业信息化的发展和提升。本文工作和主要贡献包括以下几点：（1）从云计算和云存储的理论和模型出发，面向大规模的数据密集应用，提供面向用户透明的异构数据集成和访问接口服务；（2）通过将云计算环境下异构数据集成所需要的各种基本服务分层的组织起来，为用户将现有异构数据应用向云计算环境的迁移与集成提供一种更高级的抽象服务，并可以将用户的异构数据应用无缝的映射为云计算环境下统一的数据服务和行为；（3）根据用户需要，提供云计算环境下各种业务应用数据的集中管理和统一处理，实现异构数据统一的检索与处理，以及业务应用所需的异构数据之间的实质性关联与映射；（4）本文提出的模型能够实现对云计算环境中各种关系型以及非关系型异构数据的智能集成，满足用户高并发、高负载、高速处理海量数据的复杂多表关联查询请求。因此，集团企业通过数据集成能够获取业务所需的及时且准确的信息，帮助企业进行预先控制和集成管理。集团企业实现了高度的数据集成，才能够真正对成员企业和个人进行有效监控，保证业务执行的及时和准确，提高企业的服务质量和工作效率，促进人力资源的统一调配，充分发挥企业海量数据的商业价值。
[Abstract]:In the current economic environment of our country, group enterprises play an important role. Large enterprises often have complex organizational structure, multi-level subordinate branches, numerous personnel and complex subordinate relationship. The business is distributed in many industries, the management region is wide, the management mode is diverse, and needs the omni-directional cooperative work and so on. A large amount of information of the group enterprise is scattered in different subsidiary company, different department, personnel and so on. The source of information is complex, the amount of information is very large, and the integration is very poor, which brings great challenges to the decision of enterprise production management. With the development of cloud computing technology, enterprise management applications to the cloud computing environment migration and deployment background, group enterprises urgently need through multi-data sources and multi-data center collaboration. Implement the integration of distributed and heterogeneous data in existing different information systems. Data integration can simplify the business process and work cost of enterprises and achieve the purpose of obtaining business benefits from data, so data integration has become a strategic work of enterprises. Through data integration, distributed data can be shared by group members, which makes it possible to develop new business, supervise group leaders and guard against risks. This paper proposes a heterogeneous data integration model based on ontology in cloud computing. The model is based on ontology theory and is constructed according to the characteristics of cloud environment. It supports data integration of traditional storage and cloud storage. To meet the needs of enterprise users for high concurrent access and efficient storage of mass data. At the same time, it also needs to meet the needs of high scalability and high availability of storage data and consistency of data transactions in order to complete the deployment and migration of group enterprise information applications on the cloud. The main contributions of this paper include the following points: 1) from the theory and model of cloud computing and cloud storage, oriented to large-scale data intensive applications. Provide user transparent heterogeneous data integration and access interface services; By organizing the various basic services required for heterogeneous data integration in cloud computing environments. It provides a more advanced abstract service for users to migrate and integrate existing heterogeneous data applications to cloud computing environment. The heterogeneous data application of users can be seamlessly mapped to uniform data services and behaviors in cloud computing environment. According to the needs of users, it provides centralized management and unified processing of all kinds of business application data in cloud computing environment, and realizes the unified retrieval and processing of heterogeneous data. And the substantial association and mapping between the heterogeneous data needed by the business application; 4) the model proposed in this paper can realize the intelligent integration of relational and non-relational heterogeneous data in cloud computing environment, and meet the high concurrency and high load of users. High-speed processing of complex multi-table association query requests for massive data. Therefore, the group enterprise can obtain the timely and accurate information needed by the business through data integration, which helps the enterprise to carry on the advance control and the integration management. The group enterprise realizes the high degree of data integration. In order to truly monitor the member enterprises and individuals effectively, ensure the timely and accurate implementation of business, improve the service quality and work efficiency of enterprises, promote the unified deployment of human resources. Give full play to the business value of massive data.
【学位授予单位】：天津大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：TP3;F270.7;F276.4

【参考文献】