空间高效的分布式数据存储方案

Scheme of high space efficiency in distributed storage

针对分布式数据存储中空间效率低、计算复杂度高等问题,基于Jordan矩阵和拉格朗日差值公式,提出了一种一般访问结构上高效的分布式数据存储方案。方案是计算安全的,空间利用率与理论安全的方案相比提高了m2倍,每个存储服务器只需维护长度很短的秘密份额,就可以实现大数据的分布式存储。在数据存储过程中,存储服务器根据双线性对的性质计算并贡献影子份额,确保秘密份额的安全性。方案具有可公开验证性,有效防止了数据分发者与存储服务器的欺骗。最后对方案的正确性、安全性、拓展性、空间效率等进行分析,表明方案在分布式数据安全存储中具有很好的应用前景。

Focusing on the problem of low space efficiency and high computational complexity in distributed network,this paper proposed a scheme of high space efficiency in distributed data storage on general access structure based on the theory of Jordan matrix and the formulary of Lagrange differential. This scheme was computational secure,which improved the efficiency of m2 times compared to those which were theoretical secure. Each storage server could share a long secret with each of them keeping a short share. In the process of data storing,storage server just computed and contributed the shadow according to the theory of bilinear pairing,assuring the safety of the share. The scheme was publicly verifiable,so cheating between secret distributer and storage server was avoided. In the end,it analyzed the validity,security,expansibility and space efficiency of the scheme. The result indicates that the scheme can be of good use in secure distributed storage.