基于多项式回归预取技术缓存管理的研究

Research on Technical Cache Management Prefetching Based on Polynomial Regression

Deep Web的信息资源非常丰富,涉及的领域也极其广泛,因此Deep Web的访问率也极高.随着时代的进步和发展,网络已经离不开人们的生活,缓解Web访问延迟也越来越重要.正是由于Deep Web的这些特点,缓解Deep Web的访问延迟对增加整个Web的访问速度是有重大意义的.目前缓解访问延迟采用的主要方法有预取技术和缓存技术,由此就会涉及到缓存管理的问题.本文主要是对Deep Web中采用多项式回归预取技术缓存管理的研究,从缓存的位置,存储管理和缓存替换策略等几个方面进行研究,使用户的访问速度得以提高,使预取的性能在准确率和查全率两个指标上达到满意的标准.

Deep Web information resources are very rich and extremely broad,and the covering areas of Deep Web access rate is also very high.With the progress and development,the network has been inseparable from people＇s lives,and to alleviate the Web access latency is becoming more and more important.The methods to ease the Deep Web access latency is mainly the prefetching and caching technologies,which will be related to the cache management problem.In this paper,the Deep Web using polynomial regression prefetching cache management,from the aspects of the location of the cache,memory management and cache replacement strategies,so that the user＇s access speed can be improved,so that the performance indicators of prefetching accuracy and recall rate can reach a satisfactory standard.