Dynamic remote data auditing for securing big data storage in cloud computing / Mehdi Sookhak

Sookhak, Mehdi (2015) Dynamic remote data auditing for securing big data storage in cloud computing / Mehdi Sookhak. PhD thesis, University of Malaya.

Preview

PDF (Full Text) Download (12Mb) | Preview

Abstract

Nowadays, organizations produce a huge amount of sensitive data, such as personal information, financial data, and electronic health records. Consequently, the amount of digital data produced has increased correspondingly and often overwhelmed the data storage capacity of many organizations. The management of such a large amount of data in local storage system is difficult and incurs high expenses because of high-capacity storage systems needed and the expert personnel to manage them. Although the cost of storage hardware has tremendously decreased in recent years, about 75% of the total ownership cost is still assigned tomanage data storage. It is not surprising, therefore, that cloud computing is now embraced as a key technology to provide a convenient, on-demand network access to a shared pool of configurable computing resources, and demands minimum service provider interaction or management effort. Organizations now have an option to outsource their data to cloud storage to decrease the burden on local data storage and also to reduce maintenance cost. Although the cloud offers tangible benefits to data owners, outsourcing data to a remote server and delegating management of data to an untrusted cloud service provider, can lead to loss of physical control over the data. To the clients, the cloud is inherently neither secure nor reliable and this poses new challenges to the confidentiality, integrity, and availability of data in cloud computing. Without a local copy of the data, traditional integrity verification techniques such as hash functions and signatures are inapplicable in the cloud storage. Also, it is impossible to download a large-size file from the cloud storage. The situation is made worse when users access data using their mobile devices. In this context, a more efficient technique is required to remotely verify the integrity of the outsourced data in the cloud. In this research, a new remote data auditing method is proposed for securing data storage in cloud computing based on an algebraic signature. This signature allows the auditor to check data possession in cloud storage, and this incurs fewer computational overheads on the auditor and server in comparison to Homomorphic cryptosystem. Moreover, a new data structure – Divide and Conquer Table (D&CT) – is designed to efficiently update the outsourced data dynamically by performing insert, append, delete, and modify operations by the data owner. Furthermore, the proposed method is implemented in the real environment to prove the security, justify the performance of our method, and compare with the most familiar and the stat-of-the-art data auditing methods on the basis of computation and communication cost. It is found that by employing the proposed RDA method the computational and communication costs of data integrity is reduced. D&CT data structure reduces the computation cost of data update for normal and large-scale files markedly. Hence, the proposed RDA provides an efficient and secure solution for mobile cloud computing.

Actions (For repository staff only : Login required)