电子学报 ›› 2009, Vol. 37 ›› Issue (2): 318-324.

• 论文 • 上一篇    下一篇

一种高效的累进式空间连接查询处理算法

唐桂芬1, 杨伟锋2, 黄双临1, 李 炜1   

  1. 1. 解放军61081部队,北京 100094;2. 空军指挥学院,北京 100045
  • 收稿日期:2007-01-30 修回日期:2008-09-01 出版日期:2009-02-25 发布日期:2009-02-25

An Efficient Progressive Spatial Join Query Processing Algorithm

TANG Gui-fen1, YANG Wei-feng2, HUANG Shuang-lin1, LI Wei1   

  1. 1. 61081 Troops of PLA,Beijing 100094,China;2. Air Force Command College,Beijing 100045,China
  • Received:2007-01-30 Revised:2008-09-01 Online:2009-02-25 Published:2009-02-25

摘要: 累进式空间连接查询广泛应用于空间数据集成、在线空间聚集查询系统中.本文提出一种高效的累进式空间连接查询处理算法,在现有累进式等值连接查询处理框架之上增加空间连接处理策略,使之适于处理空间连接查询;提出一种基于驻留度的动态同步替换策略处理内存溢出,提高了驻留内存空间对象的利用率;还提出了一种基于BEA的不完全连接查询处理算法有效减少磁盘连接计算冗余的I/O和CPU计算代价.实验表明,所提出的算法明显优于现有累进式空间连接查询处理算法.

关键词: 累进式空间连接, 替换策略, 不完全连接

Abstract: Progressive spatial Join query can be extensively applied in spatial data integration,online spatial aggregation query etc.systems.This paper proposes an efficient progressive spatial join query algorithm.We add spatial query strategy over existing progressive join query architecture,so it can be applied to spatial join query.We present a dynamic concurrent flush policy based on resident degree to process memory overflow,which makes memory-join phase more efficiently.We also propose an incomplete join query algorithm based on BEA,which reduces redundant I/O and CPU cost in disk join phase.Extensive experiments prove that our technique delivers results significantly faster than the previous methods.

Key words: progressive spatial Join, flush policy, incomplete join

中图分类号: