一种面向Web内容分发的缓存替换策略

发布时间：2018-04-26 22:27

本文选题：Web缓存 + 缓存替换　；参考：《高技术通讯》2013年07期

【摘要】：研究了Web内容分发和对Web服务性能有重要影响的缓存替换机制。考虑到当前的缓存替换策略主要采用基于频度与本地局部性的替换基准,而且通过对实际Web数据访问情况的分析发现访问间隔的变化率对命中率的影响具有更高的准确性,提出了一种基于访问密度与大小混合的缓存替换策略。该策略通过统计近期缓存对象的平均访问间隔,结合该对象的字节大小进行缓存替换,并分别在固定对象数目空间与对象字节空间上进行对比实验,实验结果表明,该策略比最近最少使用(LRU)和最近频繁使用(LFU)算法提升3%~5%的命中率,比空间与频度混合贪心(GDSF)算法提升5%~8%的字节命中率。
[Abstract]:This paper studies the Web content distribution and cache replacement mechanism which has an important impact on the performance of Web services. Considering that the current cache replacement strategy mainly adopts the replacement benchmark based on frequency and local characteristics, and through the analysis of the actual Web data access, it is found that the impact of the rate of access interval on the hit rate is more accurate. A cache replacement strategy based on the mixing of access density and size is proposed. By counting the average access interval of the recently cached object and combining the byte size of the object, the strategy performs cache replacement, and compares the number of objects with the object byte space. The experimental results show that, Compared with the least recently used LRUs algorithm and the recently frequently used LFU algorithm, the strategy increases the hit ratio by 3% and 8% compared with the space and frequency mixed greedy GDSFs algorithm.
【作者单位】：哈尔滨工业大学网络与信息安全研究中心;
【基金】：863计划(2011AA010705) 973计划(2011CB302605) 国家自然科学基金(61173145,60203021)资助项目
【分类号】：TP393.09

【相似文献】