混合云存储中网络稀疏大数据渗透迁移算法

Network sparse big data infiltration migration algorithm in hybrid cloud storage

当前方法没有考虑到特殊样本数据筛查的问题,导致数据渗透迁移完整度不高,所用时间较长,为此提出一种混合云存储中网络稀疏大数据渗透迁移算法。在主成分分析算法中引入信息熵的思想,对网络稀疏大数据进行降维处理,将降维结果与信息熵进行结合;筛选网络稀疏大数据的特征,将网络稀疏大数据中存在的无用特征进行剔除;计算网络稀疏大数据的敏感度、时间长度和访问频率,通过以上3个迁移因子构建网络稀疏大数据重组模型,实现混合云存储中网络稀疏大数据的渗透迁移。实验结果表明,提出算法在数据迁移完整性与效率方面明显优于传统方法,说明所提算法能够实现对网络稀疏大数据的有效迁移,为相关研究提供一定参考价值。

In view of the fact that the current method does not consider the problem of special sample data screening,the completeness of data penetration and migration is not high and the time taken is long,a network sparse big data penetration migration algorithm in hybrid cloud storage was proposed.The idea of information entropy was introduced in the principal component analysis(PCA)algorithm to reduce the dimensionality of network sparse big data,and the dimensionality reduction results were combined with information entropy.The characteristics of the network sparse big data were screened and the network sparse big data were filtered.Useless features in the data were eliminated.The sensitivity,time length and access frequency of network sparse big data were calculated.The above three migration factors were used to construct a network sparse big data reorganization model to achieve the penetration and migration of network sparse big data in hybrid cloud storage.Experimental results show that the proposed algorithm is significantly superior to traditional methods in terms of data migration completeness and efficiency,indicating that the proposed algorithm can effectively migrate sparse large data on the network and provide a certain reference value for related research.