Optimizing Data Allocation for Distributed Databases on Intranet by Ontology Schema

• Main Authors: Tzung-Liang Hung, 洪宗良
• Other Authors: Jau-Ji Shen
• Format: Others
• Language: en_US
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/39368252913637698405

碩士 === 朝陽科技大學 === 資訊管理系碩士班 === 91 === Along with the development of network and database technology, distributed databases are widely used in most of the companies. For the sake of increasing the efficiency of data processing and reducing the cost of data transmission in the distributed databases, the data should be placed in the most appropriate server sites. Most of previous researches assume that the database design is already there and then the data allocation problem is investigated on the existent distributed databases by analyzing the system operation history, such as query frequency and data affinity. In this thesis, an ontology schema will be proposed to analyze and represent the data requirements for a new distributed database system construction, and then two data allocation techniques are proposed to minimize the total transmission cost between server sites with the consideration of usage. In addition, the total transmission cost of a distributed database depends on not only the way of data allocation, but also the intranet topological design. Most of researches were addressed on the topological design problem on particular network architectures and used heuristic algorithms to achieve optimal network connections. In this thesis, a method for intranet design is proposed by establishing an ontology schema for the organizational data, and then sketching an intranet topology to achieve a deterministic solution. An intranet based on this method can not only fulfill the requirements of the enterprise information systems but also provide the minimum total transmission cost.