首页 | 本学科首页   官方微博 | 高级检索  
     检索      

联机分析查询处理中的一种聚集算法
引用本文:蒋旭东,冯建华,周立柱.联机分析查询处理中的一种聚集算法[J].软件学报,2002,13(1):65-70.
作者姓名:蒋旭东  冯建华  周立柱
作者单位:清华大学,计算机科学与技术系,软件研究所,北京,100084
基金项目:国家重点基础研究发展规划973资助项目(G1998030414)
摘    要:联机分析处理(online analytical processing,简称OLAP)查询是涉及大量数据的即席复杂查询,从SQL(structured query language)角度来看,这些查询通常都包含多表连接和分组聚集操作.从OLAP查询处理角度出发,提出一种新的基于排序的聚集查询算法MuSA(sort-based aggregation with multi-table join).该方法充分考虑到数据仓库星型模式的特点,将聚集操作和新的多表连接算法MJoin相结合,排序时采用

关 键 词:数据仓库  OLAP(online  analytical  processing)查询  多表连接  聚集查询  星型模式
文章编号:1000-9825/2002/13(01)0065-06
收稿时间:4/4/2000 12:00:00 AM
修稿时间:2000年4月4日

A Novel Aggregation Algorithm for Online Analytical Processing Query Evaluation
JIANG Xu-dong,FENG Jian-hua and ZHOU Li-zhu.A Novel Aggregation Algorithm for Online Analytical Processing Query Evaluation[J].Journal of Software,2002,13(1):65-70.
Authors:JIANG Xu-dong  FENG Jian-hua and ZHOU Li-zhu
Abstract:The OLAP (online analytical processing query) queries are ad-hoc, complex queries, as expressed in SQL, these queries include multi-table join and aggregate operation. In this paper, a novel sorting based aggregation algorithm, MuSA (sort-based aggregation with multi-table join), is given for OLAP query evaluation. In this algorithm, by taking the characteristics of star schema into consideration, the aggregation operation is combined with a novel multi-table join algorithm, MJoin, and the key words mapping technique is used to compress the sorting key which can obviously speed up sorting.Further by esting the group number of query result,the proper sorting methods which can optimize the algorithm for different aggregation queries be chosen.Asbeing illustrated by the experimental result,compared with original methods for aggregation query evaluation,theperfmance of the new algorithm can be improved dramatically.
Keywords:data warehouse  OLAP (online analytical processing) query  multi-table join  aggregation query  star schema
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《软件学报》浏览原始摘要信息
点击此处可从《软件学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号