摘要
如何在高速发展的网络环境下构建大规模、高性能、高可靠性及高可扩展性的分布式存储系统,是分布式存储技术面临的新发展课题。利用Shamir门限方案分割秘密方法的重要思想对共享数据进行分割,并将得到不同的份额再根据模糊数学中的隶属度来构造出子数据集,而后将子数据集分发给各参与者进行存储,根据用户的需要进行数据恢复。核心是把分割生成的子数据扩充生成子数据集,以此来增强数据恢复的可靠性。
Distributed data storage system have the ability to accommodate great number of users, and to deal with massive storage data distributed in the Internet environment, and provide high data availability and high performance of storage service. How to build up this data storage system with good scalability, efficiency, and reliability is a valuable and urgent job in the field of distributed storage technology. A new data distribution algorithm is proposed in this paper, in which Shamir＇s threshold scheme is used to divide the data, and then the Fuzzy data sets are constructed from the sub-share that is segmented from the original data. The sub-shares are distributed to every participant for storage at the terminal system, and also the original data could be restored according to users＇ requirements. The main contribution of this dissertation is to strengthen the reliability of data restoration by adding redundancy to the sub-shares.
出处
《信息安全与通信保密》
2010年第8期76-78,共3页
Information Security and Communications Privacy
关键词
分布式存储
秘密共享
门限方案
隶属度
distributed storage
secret sharing
threshold scheme
membership
