Research On Integrity Auditing And Repair Mechanism For Big Data In Cloud

With the rapid development of big data and cloud computing,and in today's growing attention to them,how to guarantee the data security of big data under the cloud environment has begun to become more and more important.According to the characteristics of integrity,confidentiality,availability,and non-repudiation required for data security,data security problems in big data under cloud environment mainly include data integrity destruction,data leakage,data inaccessibility and illegal use of data.We mainly focus on the big data auditing and repair mechanism in cloud computing environment,starting from the integrity verification of big data,the repair of big data,and the dynamic operation of data.The work and innovations of this thesis include the following sections:1.We have deeply analyzed and improved a schema proposed by Le et al.for encrypting cloud data,which support integrity verification.In this schema,an attacker can intercept data and tamper with the data during the data upload phase.We design an improved data upload method to enable users to be notified after the data has been tampered with.Aiming at the problem that the TPA can get information about user data by carrying out the response information returned from the storage node at the audit stage,we modify he composition structure of the response content sent from the cloud server to the TPA.In addition,the secure channel used in the original scheme does not exist in reality.So,we add a mutual authentication process for key distribution.2.A scheme named Public Integrity Auditing for Minimum Storage Regenerating Code Based Cloud Storage,PAMSR for short,is proposed.We perform security analysis and performance for the scheme.The scheme innovatively combines the minimum storage regeneration code and the TPA auditing model,and takes advantage of the low storage overhead of the MSR code.The codes are stored in multiple server nodes based on the storage structure of the Merkle hash tree,allowing auditors to check the integrity of data stored across multiple servers,and identify the location of damaged data in a round when data corruption is detected.The privacy of the data is protected during the integrity auditing.This scheme supports randomly selecting nodes from the remaining surviving nodes to repair the damaged node which is located.The scheme also supports the modification and deletion of stored data.3.A scheme named Identity-Based Dynamic Public Auditing in Clouds,IBDPA for short,which supports the batch auditing operation of data by third party auditor.In this schema,after the data owner and its authorized proxy user store the data on the server,whether the client is the data owner or the proxy user is identified based on the identity,which avoids the use of complex encryption certificates.The solution supports dynamic updating of data by any legitimate user.We improve the efficiency of data updating by using Hash Double Linked List and Location Array at the TPA side.Besides,the scheme supports dynamic updating of data by any legitimate user.