计算机工程与应用 ›› 2013, Vol. 49 ›› Issue (11): 110-113.

• 数据库、数据挖掘、机器学习 • 上一篇    下一篇

区间树在DDM区域匹配中的应用

尚福华,张海波,解红涛   

  1. 东北石油大学 计算机与信息技术学院,黑龙江 大庆 163318
  • 出版日期:2013-06-01 发布日期:2013-06-14

Application of interval-tree in region matching for DDM

SHANG Fuhua, ZHANG Haibo, XIE Hongtao   

  1. School of Computer and Information Technology, Northeast Petroleum University, Daqing, Heilongjiang 163318, China
  • Online:2013-06-01 Published:2013-06-14

摘要: 数据分发管理(DDM)是降低网络冗余数据的有效手段,区域匹配算法又是数据分发管理实现的关键。当前的多种匹配算法如直接匹配法、网格法、排序法等效率都不够理想,或者过滤效果不佳,或者耗时较长。通过对数据过滤机制的深入研究,提出了基于区间树的区域匹配算法——ITBM算法,该算法将范围的上下界映射到一个区间内,使用区间树来存储区域范围,通过对区间树的直接操作来完成匹配工作。结果表明,ITBM算法大大减少了匹配计算的时间,有效地减少了动态DDM的维护开销。

关键词: 数据分发管理, 区域匹配, 区间树

Abstract: Data Distributed Management(DDM) is the effective method to reduce network redundant data, region matching algorithm is the key of data distributed management. The current variety of matching algorithms such as direct matching method, the grid method, sorting method are insufficient ideal because of the poor filtration or long time-consuming. Through the fully research of data filtering mechanism, the region matching algorithm based on interval-tree—ITBM is proposed, which is mapped range to an interval, uses the interval trees to  store the area range, through the direct operation of interval-tree to complete matching work. The results show that ITBM can greatly reduce the time of matching calculations, effectively save the cost of matching process of dynamic DDM.

Key words: data distributed management, region matching, interval-tree