期刊文献+

基于并行处理的NN数据检索方法

A method for nearest-neighbor search based on parallel processing
下载PDF
导出
摘要 利用已有的VA-File索引结构,提出一种并行的NN检索方法,通过并行处理解决多维检索中CPU占用过多而影响检索效率的问题。同时,对本并行方法与非并行检索算法进行了比较。 One parallel processing method based on the exiting VA-File index structure is proposed in this paper, which can solve the CPU costs and improve the performance of NN search. Moreover, a fast evaluation between parallel method and non-parallel method is given.
作者 郭天辉
出处 《纺织高校基础科学学报》 CAS 2003年第2期165-169,共5页 Basic Sciences Journal of Textile Universities
关键词 多维索引 最邻近检索 并行处理 NN数据检索方法 数据库 并行性能 multi-dimensional index NN search parallel processing
  • 相关文献

参考文献5

  • 1VOLKER Gaede, OLIVER Gunther. Multidimensional access methods[J]. ACM Computer Surveys, 1998,30(2) : 170-231.
  • 2WEBER R,BLOTT S. An approximation based data structure for similarity search[z]. Technical Report 24; ESPRI Tproject HERMES (No. 9141), 1997.
  • 3ROGER Weber, SCHEK Hans-J. A quantitative analysis and performance study for similarity-search methods in high-dimensional spaces[A]. Proc 24th Int Conf VLDB[C]. 1998. 194-205.
  • 4YASUSHI Sakurai, MASATOSHI Yoshikawa. The A-tree: An index structure for high-dimensional spaces using relative approximation[A]. Proc 26th ht Cold VLDB[C]. 2000. 516-526.
  • 5WEBER R, BOHM K. Trading quality for time with nearest-neighbor search[J] . Lecture Notes in Computer Science,2000,1777: 21-35.

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部