طرح هزینه ای موثر حسابرسی پویا برای ذخیره سازی برون سپاری در محیط ابری A Cost Effective Dynamic Auditing Scheme for Outsourced Data Storage in Cloud Environment

I. INTRODUCTION In recent years, cloud computing is becoming more and more popular in both academic and commercial world. Cloud computing offers resources such as computing power, storage, online applications and networking infrastructures shared as services over the Internet[1]. Cloud storage is an important service of cloud computing, which allows data owners (DO) to outsource their data from their local computing systems to the remote cloud. It provides low-cost, scalable, locationindependent platform to store and manage users’ data. Therefore naturally more and more data owners start to use cloud storage services. Cloud storage service offers a lot of advantages and possibilities and it takes the storage management and maintenance from the hands of data owners, but it also introduces new challenges and security threats from the external sources. Security threat can also arise from inside the cloud. There are several motivations for CSP to behave unfaithfully towards the cloud users. Therefore, the owners need to be certain that the data are correctly stored in the cloud. The owner should be notified promptly if any problems regarding to owner’s data occur. Moreover in cloud storage system, none of the sides, namely data owner and cloud storage provider, is guaranteed to provide the auditing report in and unbiased manner. Therefore third-party auditing is an intuitive selection for cloud storage auditing. The efficient and dynamic auditing method utilizes a dynamic hash table data structure that allows the auditor to perform dynamic data operations effectively with minimum computational and communicational overhead on the TPA and cloud server. This paper is organized in to following sections. Section 2 reviews few of the existing state of art techniques and section 3 briefs the system model of the proposed auditing scheme. Section 4 and 5 presents the protocol of the proposed auditing scheme and analyses its performance. Finally section 6 gives the conclusion and future work of this paper. II. RELATED WORK In latest years, there has been an exhaustive research on integrity auditing of cloud data storage. Ateniese et al. [3] were the first to propose a novel approach for public auditing in their provable data possession model. They used RSA based homomorphic linear authenticator scheme for auditing the outsourced data. Their scheme is lightweight with low overhead on the server but this scheme can handle only the static storage files which will not be suitable for dynamic operations. Tan, Shuang, et al.[4] provided a dynamic version of audit supporting batch auditing through which only required data blocks can be verified based on the request from the data owner. However only limited number of queries can be handled and does not support public auditing. Erway et al.[5] introduced DPDP based on authenticated skip lists and RSA tree. DPDP provides dynamic and improved detection. Due to the various processes running in the cloud, this method produces huge computation cost at the server side. Juels et.al.[6] introduced a POR protocol in which the verifier stores only a single key and requires access to a small portion of the file for its auditing. This scheme takes lot of preprocessing and encoding of the data file before the storage in CSP which causes high computational overhead. Shacham et.al [7] provided an improved POR scheme based on the BLS signatures which reduces the communication cost but does not support dynamic operations. The dynamic updating of data is a critical feature of data auditing protocol.