基于数据差异的连续数据保护恢复算法  

RM-CBDD:Recovery Method for CDP Based on Data Discrepancies

作　　者：王超[1] 李战怀[1] 张小芳[1] 侯利曼[1] 

机构地区：[1]西北工业大学计算机学院,西安710072

出　　处：《计算机学报》2013年第11期2303-2315,共13页Chinese Journal of Computers

基　　金：国家"九七三"重点基础研究发展规划课题(2012CB316203);国家自然科学基金(61033007/F020204)资助~~

摘　　要：连续数据保护系统在进行数据恢复时,首要任务是从历史时刻中快速识别出可恢复时刻点,总恢复时间通常与恢复时刻点识别过程中所检测的历史数据版本成正比.然而,基本数据恢复方法的恢复效率低,无法适应现代存储系统对可用性和可靠性的要求,恢复时间和数据损失之间的矛盾日益突出.通过对邻近算法的改进和完善,提出了一种支持多间隙复杂情况的恢复算法——RM-CBDD.RM-CBDD通过分析并消除恢复起止时刻之间两种类型的差异数据实现恢复.实验结果表明,在多间隙复杂情况下,RM-CBDD算法的恢复效率明显优于基本方法和WDRS算法,有效降低了二分探查最佳恢复时刻点的时间开销.When the continuous data protection （CDP） system performs data recovery after data corruption, the primary task is to quickly determine a recovery point that provides a clean copy of the data. The total recovery time is proportional to the number of historical copies examined in the process of recovery point identification. The high availability and high reliability of computer- ized data has a raise requirement for modern storage systems. But the recovery efficiency of basic recovery method for block-level continuous data protection is low, and the contradiction between the recovery speed and data loss become increasingly prominent. After improve and refine the neighboring point data recovery method, this paper presents a recovery method for CDP based on data discrepancies （RM-CBDD）. In multi-gap situations, RM-CBDD performs data recovery by eliminating the data discrepancies between starting and ending points. Experimental evaluation demonstrates that the recovery efficiency of RM-CBDD is significantly better than basic method and WDRS, especially the binary search for optimal recovery point identification.

关 键 词：连续数据保护 数据恢复 数据差异 块数据 可用性 

分 类 号：TP333[自动化与计算机技术—计算机系统结构]