Description

Cloud storage services have become commerciallypopular due to their overwhelming advantages. To provideubiquitous always-on access, a cloud service provider (CSP)maintains multiple replicas for each piece of data on geographically distributed servers. A key problem of using the replicationtechnique in clouds is that it is very expensive to achieve strongconsistency on a worldwide scale. In this paper, we first presenta novel consistency as a service (CaaS) model, which consistsof a large data cloud and multiple small audit clouds. In theCaaS model, a data cloud is maintained by a CSP, and a groupof users that constitute an audit cloud can verify whether thedata cloud provides the promised level of consistency or not. Wepropose a two-level auditing architecture, which only requires aloosely synchronized clock in the audit cloud. Then, we designalgorithms to quantify the severity of violations with two metrics:the commonality of violations, and the staleness of the value ofa read. Finally, we devise a heuristic auditing strategy (HAS)to reveal as many violations as possible. Extensive experimentswere performed using a combination of simulations and realcloud deployments to validate HAS.