数据仓库性能优化之索引和物化视图耦合方法  

Coupling method of indexes and materialized views about data warehouse performance optimization

作　　者：马莹莹[1] 戴牡红[1] 

机构地区：[1]湖南大学软件学院,长沙410082

出　　处：《计算机应用研究》2013年第3期835-837,共3页Application Research of Computers

基　　金：湖南省自然科学基金资助项目(2011FJ3034)

摘　　要：为了进一步提高数据仓库的性能,通过分析数据仓库中性能优化技术的特点,提出了索引和物化视图耦合的性能优化技术。通过数据挖掘自动选择候选索引和物化视图,减少查询的扫描范围;然后研究在物化视图上建立索引的空间高效存储方法,以提高查询速率;最后利用成本模型对耦合情况进行分析,验证了耦合方法可以极大提高单一索引查询或者物化视图的性能。In order to further improve the performance of the data warehouse, by analyzing the characteristics of the data ware- house performance optimization techniques, this paper presented a performance optimization technique which was coupling in- dexes and materialized views. Firstly, it selected candidate indexes and materialized views automatically by data mining, and then researched the method of space efficient storage that indexing on materialized views. Lastly, experimental results verify that the coupling method can greatly raise the properties of a single index or a materialized view, by using the cost model to an- alyze the coupling.

关 键 词：索引 耦合 物化视图 成本模型 数据仓库 

分 类 号：TP311.132[自动化与计算机技术—计算机软件与理论]