基于半连接的分布式数据库查询优化算法探讨  

Optimization algorithm of searching a semi-connected distributive database

作　　者：杨旭超[1,2] 

机构地区：[1]杭州电子科技大学,浙江杭州310018 [2]浙江经济职业技术学院

出　　处：《计算机时代》2012年第2期16-19,共4页Computer Era

摘　　要：采用连接操作实现分布式数据查询时,不仅代价较高而且费时。若利用半连接查询操作取代连接操作,可以通过缩减操作数获得查询优化。而两次半连接对接算法可以更大限度地缩减操作数,实现起来更为简单、高效。针对多节点的数据查询,提出了通过最小生成树算法生成并行的连接序对,并由两次半连接对接算法进行查询优化处理的算法。应用该算法可以有效利用多节点的并行性缩短系统的查询响应时间,降低系统的总开销。算法在海量信息查询中具有实用价值。Searching a distributive database through connection not only costs much but is also time-consuming. Using a semi-connection search algorithm instead could attain a better result by reducing amount of operation, while a double semi-connection search algorithm may reduce more operation and is easier and more effective to realize. Focusing on Aiming to data search based on multi-node, it is proposed in this paper a connection search process through the minimum spanning tree algorithm which generates a parallel connection sequence, based on the optimum process of double semi-connection algorithm. It may reduce system response time of search lower the overall cost of system through utilizing the parallel character of algorithm. This algorithm is valuable in searching information in a huge distributive database.

关 键 词：分布式数据库 查询优化 缩减 代价 

分 类 号：G301[文化科学]