SecuredDW: A Homomorphic Schema to Securely Hosting Data Warehouse in the Cloud

Abstract

Currently, cloud computing has become the most popular technologies in the area of IT enterprise. It has many advantages such as computing power, storage, network and software as a service. Moreover, many other benefits have made it attractive. In fact, it is easy to deploy, its technical infrastructure is adaptable to the volume of business activity and its cost is relative to consumption. Whereas building a data warehouse typically necessitates an important initial investment, with the cloud pay-as-you-go paradigm, BI system can benefit from this new technology. However, cloud computing brings its own risks in terms of security. For this purpose, before outsourcing sensitive data to the cloud, the owner must encrypt his data to keep secure. In the particular context of cloud data warehouse, privacy is of critical importance because it contains sensitive data. Cloud provider proposes traditional security solutions to ensure the confidentiality of outsourced data. Unfortunately, those solutions are not practical in the case of data warehouse anymore because they induce a heavy overhead in terms of data storage and query performance. So, a new solution must be proposed for outsourcing data warehouse to the cloud that respects its specification and swings performance and security. In this paper, we propose (SecuredDW) a new sharing schema for securing and querying a data warehouse hosted in the cloud based in a homomorphic algorithm. The integrity of data is also addressed in this paper by proposing two new signatures to verify the correctness of data sent and received from the cloud. Theoretical results show the efficiency of Secured DW in terms of privacy and performance with respect to other solutions.