面向异构数据集成的数据源能力描述框架  

A Data Source Capability Description Framework towards Heterogeneous Data Sources Integration

作　　者：唐九阳[1] 张维明[1] 修保新[1] 肖卫东[1] 

机构地区：[1]国防科技大学管理科学与工程系,长沙410073

出　　处：《计算机科学》2004年第4期170-172,188,共4页Computer Science

基　　金：国家自然科学基金(60172012)

摘　　要：异构数据源集成系统旨在为用户提供一个一致的访问接口,由于参与集成的各数据源不仅高度自治、模式各异、更新频繁,而且查询功能有各自特殊的限制,给查询处理过程中数据源定位和查询优化造成一定的困难。本文在分析异构集成系统特征和功能需求的基础上,提出一种基于KQML的数据源能力描述框架,为各数据源灵活动态的发布自身能力提供保证。进而通过形式化的规范描述刻画数据源的结构特征和行为特征,为定位查询相关数据源奠定基础,并有助于全局查询处理器对查询计划进行优化,缩减查询的搜索空间,提高查询效率。The heterogeneous data sources integration systems aim at providing a uniform interface for applications. The integrated data sources can be full-fledged databases .simple files, HTML pages or specialized data sources that posses diverse query processing capabilities, which make it difficult for the location of data sources and query optimization. This paper investigates the issues in data sources description in terms of their contents and processing capabilities, and proposes a framework to describe the capabilities of the data sources in fine detail. When a data source is registered with the mediator,the mediator uploads the capabilities of the wrapper,and smoothly integrates these capabilities into query processing, making a well foundation for the discovery of data sources and query optimization.

关 键 词：异构数据源集成系统 数据源 异构集成 计算机网络 WWW 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]