Research On Recognition Algorithm Of Damage Data In Cloud Storage

With the rapid development of cloud computing,more and more data is stored in the cloud server.As a new kind of storage service,cloud storage has the characteristics of high scalability,high reliability,low price and access anytime,anywhere,and provides a new model for the massive data storage management.However,because users host the data to the cloud service provider,the users will lose the local control of the data,so cloud storage is facing many challenges.One important issue is how users determine whether the data stored in the cloud server is corrupt.Traditional data integrity verification methods require data to be downloaded locally for damage identification.Due to the large amount of data in the cloud storage,download the data block will bring a lot of network burden,so the traditional methods and techniques are not suitable for cloud storage environment.At present,in the existing data integrity verification algorithm for cloud storage,the more excellent is based on BLS short signature algorithm.They are based on data blocks as the basic verification unit,have the characteristics of unlimited number of checksum,privacy protection when checking,low communication overhead and so on.However,due to a large number of modular exponents in the algorithm,the computational cost is very high.And it believes that third-party verifier is fully credible,does not meet the reality of cloud computing environment,especially mobile cloud computing environment.Considering the incomplete trustworthiness of the executors in the process of damage data identification in cloud storage,that is,the verifier can deceive the data owner or conspires with the cloud storage server to hide the damaged data state so that the original damaged data can be invalidated,which cannot guarantee the timely discovery of the damaged data and waste of computing resources.Based on the support of third-party verification,a dual-evidence-based data identification algorithm based on Homomorphic Verifiable Tags(HVT)is proposed to efficiently identify damaged data in cloud storage and verify the reliability of the recognition results of the third party verifier with a high probability.The data owner selects a set of suspect corrupted data block,computes a set of proofs for the data owner using the challenge,data,public key,etc.,while the cloud server computes the another challenge evidence of the suspect corrupted data,so that the data owner verifies the auditor's verification result.The algorithm not only verifies the integrity of the data,but also identifies the reliability of the verification result of the third party verifier with a sufficiently high probability and identifies the damaged data.At the same time,in the mobile environment,the energy of mobile devices,CPU and other computing resources is limited,in order to avoid the mobile auditor because of limited computing resources caused by the task of the failure of resources,resulting in waste of resources,according to the value of data to be identified density and verification task urgency,this paper designs a task scheduling algorithm based on dynamic priority for damage data recognition,fully taking into account the third party verifier's ability to perform tasks,and according to the verifier's computing resources to dynamically adjust the task of identifying tasks,reduce task loss and waste of resources due to the inability to complete identification tasks in time.In order to maximize the value of the recognition task,we take full account of the value of each data block to be identified,maximize the entire recognition system,and realize the value accumulation of task execution.