基于CORBA的异构数据库集成研究

【作者】 高鹏；

【导师】 李怀香；

【作者基本信息】 广东工业大学 ， 计算机应用技术， 2001， 硕士

【摘要】 异构数据库集成是自80年代中期以来数据库界的研究热点。许多方法被提出，如数据库网关、ODBC等，以期解决集成系统所面临的异构性等问题。但是由于没有统一标准和构件化思想，使得集成系统实现困难，管理复杂；基本处于理论研究中，难以实现商品化。 分布式对象技术始于90年代初，已经发展成为当今分布异构环境下，建立应用系统集成框架和标准构件的核心技术，并在企业资源集成、分布式系统管理、软件构件技术等方面发挥重要作用，出现了以OMG的CORBA和Microsoft的COM/DCOM为代表的应用互操作标准。CORBA在跨平台、可扩展性、可集成遗留应用等方面优于其它标准。因此，将CORBA技术引入异构数据库集成研究中将带来多方面的好处。大大提高集成系统的可管理性、可伸缩性、可维护性和可集成性，并使系统的开发和使用都变得简单。 MyIBase系统是作者研制的一个基于CORBA的分布异构数据库集成系统。该系统集成各局部数据库系统，以全局数据库的方式向用户提供查询功能，用户看到的只是一个全局数据库，实际的用户查询被系统分解为针对各子系统的子查询，子查询结果再由系统汇总返回给用户。同时，各子系统都保持自治性，具有其本地的不为全局所知的事务运行。 结合MyIBase系统的设计与实现，本文探讨了异构数据库集成研究所面临的问题及本系统基于CORBA的解决方案；并结合MyIBase系统特色阐述了基于CORBA的异构数据库集成系统的巨大优势。本课题将对这一新兴领域的研究起到一定的推动作用。更多还原

【Abstract】 Heterogeneous database integration is an important branch in database research. Many schemes were designed, such as database gateway. ODBC etc. in order to shield the heterogeneity which the integration system has to face. But all these schemes don work very well due to lacking of component thinking and standardization. It difficult to wnplement the system and manage the system. Most of the research in this field is theoretic, and hard to become commodity.Distributed Object Technology stemmed from 1990. It has had rapid development to become the core one in building system integration frames and standard components. It also play important role in the integration of enteiprise resources, the management of distiibuted system and the technology of software components. Many cooperation standards had been published, represented by 0MG CORBA and Microsoft COMIDCOM. CORBA has better performance than others in cross-platform and extensibility, it is also good at integrating existed system. Becouse of these good performance, using CORBA to integrate subsystems will make the integration system easier to manage, easier to extend, easier to maintain and easier to implement.MylBase is a CORBA-based heterogeneous database integration system developed by us. It integrates the local database systems, gives the user a consistently globe view of the local databases by Globe Database. User won feel the heterogeneity of local databases. User query is processed by the Globe Database, which decomposed the query into a set of subqueriesne for each local database. The results coming back from the local database are organized by the Globe Database and then returned to the user.According to the design and implementation of MylBase, this paper discuss the problems in heterogeneous database integration and give our CORBA-based solution. This paper also illustrate the benefits of using CORBA in the procedure of integrating by telling the excellent characteristic of MylBase. This paper will put forward theIIresearch of CORBA-bed heterogeneous dataase integhion SyStem.更多还原