Data integrity and availability verification game in untrusted cloud storage

Abstract

The recent trends towards outsourcing data to the Cloud as well as various concerns regarding data integrity and availability created an increasing interest in enabling secure Cloud data-centers. Many schemes addressing data integrity issues and complying with various requirements came to place: high scheme efficiency, stateless verification, unbounded use of queries and retrievability of data. Yet, a critical question remains: how to use these schemes efficiently, i.e. how often should data be verified. Constantly checking is a clear waste of resources but only checking at times increases risks. This paper attempts to resolve this thorny issue by formulating the data integrity check problem as a non-cooperative game and by performing an in-depth analysis on the Nash Equilibrium and the engineering implications behind. Based on our game theoretical analysis, the course of action was to anticipate the Cloud provider’s behavior; we then derive the minimum verification resource requirement, and the optimal strategy of the verifier. Finally, our game theoretical model is validated by showing correctness of the analytical results via simulation on a case study.