Enhanced security approach for outsourced data auditing in cloud storage

Abstract

Cloud Storage prevails over the entire world and makes the mankind to rely more on a number of online storage systems to back up data or for using it in real time which gives an anywhere, anytime access. Data storage is an essential aspect in this era where the data growth is exponential. Enormous sensitive data is stored into the cloud. The dynamic data refers to information that is asynchronously changed as further updates to the information becomes available. There is no assurance that the data outsourced on remote storage servers are safe and not been lost or modified. As the user s data are stored and maintained out of user s premises, these services create concerns of security vulnerabilities for all the services provided by them. Extensive research is carried out to solve the issues with the service providers and cloud security. The primary issue of security arises as the data is outsourced to remote servers and there is less control over the users data. This vulnerability causes data loss and modifications to occur. Some of the major security issues are data integrity verification, deduplication and misbehaviour detection in the cloud storage. To verify the integrity of the data, third party auditing is necessary. Conventional protocols that deal with integrity verification and auditing are not cost effective. Hence there is a need for developing cost-effective auditing protocols for integrity verification of dynamic data. The integrity verification mechanism proposed are cost effective in terms of computation process, communication transaction of messages between the entities and storage data space usage of the Data Owner (DO) to gain trust and security assurance of the data stored. Deduplication is a significant technology used to reduce bandwidth and space by removing redundant data and storing only a unique copy of them. When the data are encrypted by the users random keys producing and the encrypted data could not be matched to identify similarity and discard them from storing it. Hence there is a