摘要
针对虚拟数据库系统 (简称“VDBMS”)执行分布式查询时的网络代价问题 ,提出一种基于关系运算代价估算理论及半连接技术的分布式查询优化算法SJQA (Semi joinBasedQueryAlgorithm ) .较之最大子查询法 ,该算法加入了对子查询之间执行顺序及连接方式等方面因素的考虑 ,从而能更有效地减少分布式查询的总网络代价 .实验结果表明 ,对于带有选择运算的查询 ,SJQA算法的优化效益与查询的选择因子大小成近似反比关系 ,因此它对该类查询的优化效果是十分明显的 .
Concerning the distributed querying network cost problem of virtual database system (shortened as 'VDBMS'),an optimized distributed query algorithm SJQA( Semi-join Based Query Algorithm ) was proposed. In comparison with Max Sub-Query method,this algorithm takes into account the executing sequence of all the Sub-Queries and their way of connection,and thus efficiently reduces the total network cost. Experimental results show that the optimizing benefits from SJQA will increase as the selective factors of the query are smaller,and its optimizing effect on this kind of query is quite obvious.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2003年第11期24-29,共6页
Journal of South China University of Technology(Natural Science Edition)
关键词
虚拟数据库
分布式查询
查询优化
网络代价估算
半连接
virtual databay
distributed query
query optimization
netwotk cost evaluation
semi-join
