首页 | 本学科首页   官方微博 | 高级检索  
     

一种基于预取感知接纳策略的查询结果缓存方法
引用本文:马宏远,王斌. 一种基于预取感知接纳策略的查询结果缓存方法[J]. 计算机研究与发展, 2012, 0(Z1): 148-152
作者姓名:马宏远  王斌
作者单位:中国科学院计算技术研究所;中国科学院研究生院
基金项目:国家自然科学基金项目(60873166,61070111);国家“九七三”重点基础研究发展计划基金项目(2007CB311103);国家“八六三”高技术研究发展计划基金项目(2006AA010105);教育部科学技术研究重点项目(109028)
摘    要:针对搜索引擎查询结果缓存问题,提出了一种基于预取感知接纳策略的查询结果缓存方法,用于提高搜索引擎检索系统性能.查询结果预取导致查询结果页码的缓存缺失率具有显著差异性,结合该特性设计预取感知接纳策略,该策略包含查询评估模型以及模型特征选择方法.在该策略基础上,设计了一种查询结果缓存方法.在该搜索引擎两个月的大规模真实用户查询日志上的实验结果表明,与传统的典型方法相比,该方法可以获得6.38%~11.99%的缓存命中率提升.

关 键 词:缓存  预取  查询结果  性能优化  接纳策略  搜索引擎  信息检索

An Approach Based on Prefetch-Aware Admission Policy for Query Results Caching
Ma Hongyuan,and Wang Bin. An Approach Based on Prefetch-Aware Admission Policy for Query Results Caching[J]. Journal of Computer Research and Development, 2012, 0(Z1): 148-152
Authors:Ma Hongyuan  and Wang Bin
Affiliation:1(Institute of Computing Technology, Chinese Academy of Science, Beijing 100190) 2(Graduate University of the Chinese Academy of Sciences, Beijing 100049)
Abstract:Query results caching is efficient techniques for Web search engines. In this paper, we present a novel approach tailored for query results caching, and the approach is based on prefetch-aware admission policy. Different page numbers have significantly different cache miss rates because of query results prefetching. We design a prefetch-aware admission policy which contains an assessment model for queries and a feature selection method. On the basis of the policy, we propose a new approach for query result caching. We then use a real large scale query logs to evaluate the approach, as well as traditional methods. Experimental results show that this approach can achieve 6.38% to 11.99% increase for all requests as compared with state-of-the-art methods.
Keywords:caching  prefetching  query results  performance optimization  admission policy  search engine  information retrieval
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号