论文部分内容阅读
云系统中的漂移数据作为大量冗余数据的一种,对其有效及时的删除能保证云存储系统的稳定与运行。传统的云系统漂移数据删除技术采用全文件静态分块策略,由于操作粒度较小,对漂移数据的删除率不高。提出一种基于垂直分层布隆过滤的云系统漂移数据删除算法,设计基于客户端/服务器的云存储系统漂移数据删除机制体系构架,采用布隆过滤技术进行数据源端处理,引入热度垂直索引热度来衡量数据块边界偏移所造成用户的访问集中热点损失,文件根据内容划分成可变长度的数据块,进行垂直分层,得到备份集中数据对象的粒度,采用奇异值分解的方式对漂移数据序列的细节信号展示,根据矩阵奇异值分解矢量的唯一性,避免一些数据块边界偏移造成的误删和漏删。仿真实验表明采用该算法进行云系统的漂移数据删除,性能较好,执行效率和精度优越于传统算法。
Drift data in the cloud system as a large amount of redundant data, effective and timely deletion can ensure the stability and operation of the cloud storage system. The traditional cloud system drift data deletion technology uses the whole file static segmentation strategy. Due to the small operation granularity, the deletion rate of the drift data is not high. This paper proposes a drifting data deletion algorithm based on vertical stratified Bloom filtering for cloud system, designs a cloud-based storage system drifting data deletion mechanism architecture based on client / server, uses Bloom filtering technology to process data source, Heat to measure the data block boundary offset caused by the user’s access to focus hot spots, the file is divided into variable length data blocks according to the content, the vertical stratification, backup granular data object size, the use of singular value decomposition of the drift The detail signal of the data sequence is displayed. According to the uniqueness of the vector of matrix singular value decomposition, the mistaken deletion and deletion of some data block boundaries are avoided. The simulation results show that the proposed algorithm can remove the drift data in the cloud system, and has better performance and better implementation efficiency and accuracy than the traditional algorithm.