摘要
为提高云存储的访问速率并降低费用,提出了一种面向费用优化的云存储缓存策略。利用几乎免费的局域网环境下的多台桌面计算机,在本地建立一个分布式文件系统,并将其作为远端云存储的缓存。进行文件读取时,首先查找其是否在缓存中,若存在则直接从缓存读取;若不存在则从远端云存储读取。采用了最近最少使用(LRU)算法进行缓存替换,将冷门数据从缓存中替换掉。以亚马逊简单存储服务(S3)作为远端的云存储服务,对原型系统进行了简单的性能测试。测试结果表明,使用了所提出的缓存策略后,在降低费用的同时能够显著提高文件读取的速度。
A cost optimization-oriented caching strategy for cloud storage was proposed to increase the access rate and reduce the cost. Nearly free desktop machines in local area network environment were utilized to build a local distributed file system, which was deployed as a data cache of remote cloud storage service. When the user read a file, it first determined whether the file was in the cache. If it was in the cache, the file could be read directly from the cache; otherwise, the file was read from remote cloud storage service. Least Recently Used（ LRU） algorithm was used for cache replacement, and the cold data were replaced. A simple performance evaluation of prototype system was accomplished using Amazon Simple Storage Service（ S3） as the remote cloud storage service. The results show that using the proposed caching strategy not only reduces cost, but also greatly improves file reading speed.
出处
《计算机应用》
CSCD
北大核心
2014年第11期3109-3111,3143,共4页
journal of Computer Applications
基金
湖南省教育厅科学研究项目(12C0121)
关键词
云存储
数据缓存
分布式文件系统
费用优化
cloud storage
data cache
distributed file system
cost optimization
