摘要
物化视图选择方法大多是静态的,违背了联机分析处理和决策支持系统的动态本质。现有的动态算法也不能实现完全的动态化,为此提出了一种数据仓库中基于聚类的动态物化视图选择算法CBD-MVS(clustering-based dynamic materialized view selection),该算法采用层次聚类技术对用户查询语句进行聚类,提出视图合并算法建立候选物化视图,利用BPUS(benefitper unit space)算法生成最终应该被物化的视图。实验结果表明该算法是有效可行的,由于采用聚类技术,实现了完全的动态化。
The materialized views selection approaches are static mostly, which greatly disobey dynamic nature of OLAP and DSS. The current dynamic materialized views selection approaches can not achieve dynamic completely. So an clustering-based dynamic algorithm for materialized view selection in data warehouse that exploits hierarchical clustering technique （CBD-MVS） is proposed, in order to determine clusters of similar queries. A view merging algorithm that builds a set of candidate views, as well as BPUS for selecting a set of views to materialize are proposed. Experimental results demonstrate its efficiency and viability. Dynamic realize completely because of exploiting clustering technique.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第15期3638-3640,3644,共4页
Computer Engineering and Design
基金
国家自然科学基金项目(60572112)
江苏省高技术基金项目(BG2007028)
江苏省六大人才高峰基金项目(07-E-025)
江苏省教育厅基金项目(06KJB120051)
关键词
数据仓库
物化视图选择
动态
聚类
视图合并
data warehouse
materialized view selection
dynamic
clustering
view merge
