A Model to Detect the Integrity Violation of Shared File in the Cloud

Abstract

Nowadays, there is an increasing trend in outsourcing data to remote cloud servers, where users store their data at specific storage space offered by Cloud service provider (CSP) with low cost. Many of cloud service provider pretend that user's data is saved securely with no data integrity violation. But as we know, anything saved in the internet environment can't be perfectly safe which reduce users trust that their files in safe situation, then they can’t store sensitive data in the cloud. So, it's vital to ensure that data are kept properly in a consistent way and data integrity is guaranteed. Surely, this advantage will be more useful when data is shared between a group of users. Some previous works treated with file integrity violation problem from the perspective of just one file without dealing with file sharing. On the other hand, some works enabled Third-Party Verifier which cause some security fears. In this work, we work on the problem of integrity of shared data over the cloud and a new model was built with dynamic data operation support. This model contains an intermediate system to detect file integrity violation by hashing the file, using SHA-256 algorithm, before saving it on the cloud and after retrieving it, then compare the two hash values. Several experiments were conducted to ensure the ability of detecting data integrity violation, different sizes of .doc files starting from 13KB, ending with 5MB were collected. On the other hand, different types of tests were conducted to evaluate the accuracy and concurrency. The evaluation process proves that our intermediate system is able to guarantee Files accuracy and detect all type of file modifications. Also, it is able to organize concurrency operations on shared file and enable just one user to modify File at the same point. As it depends on clear logic of comparing File Hash values, before storing to the cloud and after retrieving from the cloud, so there is no way to make mistakes in detecting file integrity violation and it is highly efficient to detect integrity violation of shared files.Keywords-cloud computing, integrity detection, shared file, dynamic data operations, accuracy, concurrency.