一种基于bloom-filters的半连接查询优化算法  

A semi-join query optimization algorithm based on bloom-filters

作　　者：孙中利[1] 戴玉刚[1] 刘战东[1] 

机构地区：[1]西北民族大学中国民族信息技术研究院,甘肃兰州730300

出　　处：《电子设计工程》2011年第4期1-3,共3页Electronic Design Engineering

基　　金：国家科技支撑计划资助项目(2009BAH41B00)

摘　　要：以传输费用最小为目的,提出一种新的查询优化算法。该算法以连接属性为关键字,利用半连接关系建立bloom-filters,在半连接关系间相互传送bloom-filters,从而缩减大部分不参与连接的元组,最终形成了计算结果表。通过站点间传送计算结果表来缩减连接关系,半连接的准确性比估算连接结果高,半连接查询优化算法能较准确地做出下一步的连接;新的查询优化算法能有效地得到连接操作的执行计划,从而减少了传输费用。In order to minimize the cost of transmission, this paper presents a new query optimization algorithm.The property of join was used to the keywords in this algorithm.Bloom-filters was conveied between the semi-join relationship,was established by the relationship of the semi-join.By this method ,most of the groups which was not involved in the connection were reduced. Finally,the results table was formed.By transimission the results between the sites,the connections were reduced.The accuracy result of semi-join is higher than estimated connection.The connect of the next step can be made more accurately by the semi- join query optimization algorithm.The operations of implementation plan can be obtained effectively by the new query optimization algorithm,so the transmission costs is reduced.

关 键 词：数半连接 分布式数据库 查询优化 bloom—filters 

分 类 号：TP319.9[自动化与计算机技术—计算机软件与理论]