Modeling adaptive security-aware task allocation in mobile cloud computing

Abstract

Security is one of the most important criteria in the management of cloud resources. In Mobile Cloud Computing (MCC), secure allocation of tasks remains challenging due to the limited storage, battery life and computational power of mobile devices connected to the core cloud cluster infrastructure. Secure wireless communication channels and protocols for protecting the data and information sent to the cloud, and remote access to secure cloud services, are other important problems related to task scheduling and processing in dynamic MCC. In this paper, we developed a new security-aware task allocation model strategy in Mobile Cloud Computing. In this model, we define an allocation algorithm which generates an optimal and secure configuration of communication protocols in order to meet the specific data confidentiality requirements defined by end users. Resource utilization is predicted using Machine Learning methods, and the optimal secure service for task execution is selected. We developed a simulation environment (MocSecSim) for the evaluation of the algorithms proposed in several scenarios based on the users’ requirements. The results of simulations and experiments have demonstrated that the model proposed significantly improves the level of security of calculations in comparison with a configuration where processing time and energy consumption are the main criteria for optimizing task allocation.