PDF全文下载:2014040410
张韧志, 宋喜忠
(黄淮学院 信息工程学院,河南 驻马店 463000)
摘要: 为有效减少云计算过程中不同元数据的工作量,本研究提出了一种全新的管理方案。通过改进的Chord算法和协同双层缓存机制使得元数据服务性能随着聚类上元数据服务器数量增加而增强;采用亲密算法控制客户端与服务器端的协作以便对元数据集进行缓存。基于以上方案能表现出较高性能和良好的可扩展元数据服务,实验表明本文提高的方案性能优于其它方案,更能体现出元数据访问性能与亲密值之间变化的关系。
关键词: 云计算;元数据;缓存;高性能;灵活管理
中图分类号: TP 301.6文献标志码: A
收稿日期: 20130923
基金项目: 河南省自然基金项目(122112223146);河南省科技攻关项目(13A520786).
作者简介: 张韧志(1980—),男,讲师.
A New CloudBased Metadata Management Solution
ZHANG Renzhi, SONG Xizhong
(School of Information Engineering, Huanghuai University, Zhumadian 463000,China)
Abstract: To reduce the workload of different metadata cloud computing process, this paper proposes a new management plan. The improved Chord algorithm and cooperative caching mechanism makes double metadata service performance increases with the number of clustering on data server increases; the cooperation close algorithm to control the client and server to cache metadata set. The above scheme showed high performance and good scalability of metadata service. Experiments show that the scheme outperforms the other schemes to improve, to better reflect the relationship between metadata access performance and close value.
Key words: cloud computing; metadata; cache; high performance; flexible management