基于压缩特征编码的混合云冗余数据删除算法

Hybrid Cloud Redundant Data Delete Algorithm Based on Feature Code Compression

混合云存储系统的大数据部署和管理过程中,出现大量冗余数据,需要对冗余数据合理删除,获取想要云端的数据,提高系统稳定性。传统的冗余数据删除算法会在分数阶Fourier域出现伪峰峰值,不能有效地对冗余数据进行检测滤波和删除处理,提出一种基于压缩特征码的混合云冗余数据删除算法。预测出不同时间片内混合云的任务执行期望完成时间,对混合云数据冗余主成分进行特征编码,表征为校验信息存储子集对部分冗余数据的块层结构,提高冗余数据删除性能,实现算法改进。仿真结果得出,该算法对云存储系统中冗余数据的检测性能较好,有效避免数据信息流的干扰特征造成的误删和漏删,冗余数据删除准确性高,具有较好的应用价值。

Large data deployment and management process of the mixed cloud storage system, the emergence of a large number of redundant data, need reasonable delete the redundant data, to obtain the cloud data, improve the stability of the system. Redundant data traditional deletion algorithm there will be false peaks in the fractional Fourier domain, it cannot effectively to detect and delete the redundant data filtering process, a deletion algorithm hybrid cloud redundant data compres- sion based on feature code is proposed. To predict different time slice hybrid cloud expected completion time of task execution, feature code for the mixed cloud data redundancy principal component, characterized as the check information is stored on the part of the subset of redundant data block layer structure, improve the redundant data delete performance, improved algorithm. Simulation results show that the algorithm is better on the detection performance of the redundant data in the cloud storage system, effectively avoid the interference characteristic data of information flow and leakage caused by delete redundant data, it has high accuracy, and it has good application value.