云计算环境下不可靠数据恢复方法研究

由于云计算环境下不可靠数据中包含有利用价值的信息,但直接使用可能会给计算机带来负担,所以需要对云计算环境下不可靠数据进行恢复；当前大多数方法对不可靠数据进行复写时,利用地址映射层分配新的写入地址,因此不可靠的数据在一段时间内仍存在于云存储中,在这样的系统上实现对其的保护,不需额外保存数据的更新信息;通过FLASH的带外区记录时间戳,在FTL的映射项中增加时间戳信息,在不跟踪映射表每次更新的情况下,达到数据快速恢复的目的;但这种方法对云存储系统性能产生不利影响；为此,提出一种基于张量Tucker阈值的云计算环境下不可靠数据恢复方法,首先利用云计算环境下节点自身的随机秘钥生成器产生随机会话密钥,并对不可靠数据HMAC报文鉴别码进行计算,从而实现保护和重构；在此基础上,将不可靠数据的阈值分解过程与奇异阈值方法相结合,从而得到Tncker阈值算子,实现动态的数据恢复,恢复过程中利用Tucker阈值算子与增广拉格朗日乘子方法相结合的方式选择n-秩相似张量,提出基于增广拉格朗日乘子方法的不可靠数据Tucker阈值恢复方法,完成云计算环境下不可靠数据恢复;实验证明,所提方法能够有效提高不可靠数据恢复的准确性,降低数据恢复的能耗和时间,具有较强的可行性,为该课题的应用研究提供理论依据。

Research on Unreliable Data Recovery Method in Cloud Computing Environment

Because the cloud computing environment is unreliable and the data contain valuable information, the direct use may lead to the burden on the computer, so it is necessary to recover the unreliable data in the cloud computing environment. Most of the current methods of copying unreliable data, use the address mapping layer assignment write new address, so unreliable data in a period of time still exists in the cloud storage, the protection is achieved in such a system, do not need to update the channel additional data storage information. The timestamp information is added to the mapping term of FTL through the outer zone record timestamp of FLASH, and the fast recovery of data is achieved without updating the mapping table each time. But this approach has a detrimental effect on the performance of cloud storage systems. Therefore, this paper proposes a method to calculate environment unreliable data recovery method based on tensor Tucker threshold cloud, first calculate the random secret key generator node environment generates a random session key using the cloud, and the unreliable data HMAC message authentication code is calculated, so as to realize the protection and reconstruction. On this basis, combining the unreliable data threshold decomposition process and singular threshold method, in order to get the threshold of Tncker operator, to achieve dynamic data recovery, similar n- rank tensor combined with Tucker threshold operator and the recovery process using augmented Lagrange multiplier method is chosen, no reliable data recovery method of Tucker threshold augmented Lagrange multiplier based on the proposed, completed under the cloud computing environment is not reliable data recovery. Experimental results show that the proposed method can effectively improve the accuracy of unreliable data recovery, reduce the energy consumption and time of data restoration, and has a strong feasibility. It provides a theoretical basis for the application of this topic.