Overview of Auditing Cloud Consistency

Abstract

Cloud storage services have become very popular due to their infinite advantages. To provide always-on access, a cloud service provider (CSP) maintains multiple copies for each piece of data on geographically distributed servers. A major disadvantage of using this technique in clouds is that it is very expensive to achieve strong consistency on a worldwide scale. In this system, a novel consistency as a service (CaaS) model is presented, which involves a large data cloud and many small audit clouds. In the CaaS model we are presented in our system, a data cloud is maintained by a CSP. A group of users that participate an audit cloud can verify whether the data cloud provides the promised level of consistency or not. The system proposes a two level auditing architecture, which need a loosely synchronize clock in the audit cloud. Then design algorithms to measure the severity of violations with two metrics: the commonality of violations, and the oldness value of read. Finally, heuristic auditing strategy (HAS) is devised to find out as many violations as possible. Many experiments were performed using a combination of simulations and a real cloud deployment to validate HAS.
DOI: 10.17762/ijritcc2321-8169.150111