Improving the dependability of cloud environment for hosting real time applications

Abstract

The growing use of cloud computing in various fields makes the dependability of clouds a major concern in both industry and academia, especially for real-time applications. In cloud computing, the processing is done on remote cloud nodes; therefore, the chances of errors occurring are increased because of the loose control over the remote nodes and the unexpected latency. Hence, fault tolerance in cloud computing is important to ensure the reliability and availability of real-time applications. The contribution of this paper is two-fold: first, it proposes a comprehensive framework that adopts a number of fault tolerance techniques to improve the dependability of cloud environments to host real-time applications while achieving reliability and availability requirements. Second, a fault tolerant real-time scheduling algorithm has been developed to minimize the rate of missing deadlines, makespan, and the degree of load imbalance. Keywords: Cloud computing, Dependability, Fault tolerance, Real-time applications, Cloud scheduler