【摘要】 一个成功的OLAP服务系统的体系结构必须具有足够的通用性和可扩展性,它应该能方便地与大型数据仓库系统集成,而不依赖于某个具体的DBMS软件,为此,本文提出了一个通用的、可扩展的、开放的OLAP体系结构。在此结构下,OLAP分析的数据源可以是数据仓库、多维数据库和OLTP数据库,它们可以建立在不同的DBMS上;通过语义对象的定义对用户屏蔽了不同数据源的差异,用户只需对自己熟悉的语义对象进行操作,就可获得分析结果,达到了良好的通用性。更多还原

【Abstract】 The framework of a successful OLAP Server System must b e universal and extensible, which can integrete with many big datawarehouse systems and does not dep end on special DBMS. Therefore,this paper provides an universal and extensible framework for OLAP Server System , Multidimensional database and OLTP databse which can be builded on different DBMS;And in this framework ,the defference of physical schema is shielded for users through semantic objects who can manipulate their famil iar semantic objects, and then obtain the analysis result. Therefore, this framework attains well universality.更多还原