Online allocation of cloud resources based on security satisfaction

Businesses are becoming increasingly interested in exploiting the Cloud Computing technology. However, Cloud insecurity is still among the main factors that are blocking the full migration towards this paradigm. Increasing security investments will speed up the Cloud adoption process and improve the trustworthiness of the Cloud Service Providers (CSP). Moreover, the integration of the security element into the process of resource allocation will help increase the protection of the deployed services. However, this integration requires suitable modeling of customers' security requirements and CSPs' security offerings. To this end, we propose in this paper a broker-based model for the allocation of resources in the Cloud based on service security satisfaction. The resource allocation problem is modeled as a linear optimization problem and solved using an Evolutionary Computation approach, namely, the Genetic Algorithm (GA). The objective is to maximize the global security satisfaction of users' services by placing them on the data centers that adhere the most to their security requirements. Results show that the GA achieves an acceptable approximation of the optimal solution and is computationally efficient, which makes it suitable to function in online mode and cope with the scalability of the Cloud environment.

Uncontrolled Keywords

cloud computing; cloud security; security satisfaction; resource allocation; genetic algorithm