An Energy-Efficient Scheduling Framework for Cloud Using Learning Automata

Cloud computing is an emerging paradigm which helps to realize scalability, high availability, and cost-efficiency. Various applications like financial transactions, health care system, video streaming, IoT applications need on-demand provisioning of cloud resources to assure timeliness and high availability. However, the exponential growth of data generated by these applications causes issues like guaranteeing a timely response, minimizing energy consumption while ensuring high resource utilization. Moreover, low resource utilization also causes wastage of energy. Task scheduling plays a significant role in efficient resource utilization and thus reducing energy consumption in the cloud system. In this paper, we present a learning automata-based framework and algorithm for real-time task execution in the cloud. The learning automata framework guide the scheduler in scheduling a task, realizing optimized energy consumption. The effectiveness of the proposed approach is validated by conducting a set of rigorous evaluations. Simulation results indicate the suitability of proposed algorithm against other existing algorithms.