A New Algorithm for Energy Efficient Task Scheduling Towards Optimal Green Cloud Computing

Abstract

These days the continuous use of the cloud data centers has been increased, which augmented the consumption of energy as a whole heap. The main concerns of green cloud computing are reducing and managing large amount of energy uses. In this paper, we propose a new algorithm for energy efficient task scheduling in green cloud operation. The proposed algorithm uses the Dynamic Voltage and Frequency Scaling (DVFS) and DVFS enabled Efficient energy Workflow Task Scheduling (DEWTS) in calculating the order of tasks and assigns it to the processor nodes. The algorithm helps in utilizing the idle slots of CPU without violating the constraints in the whole process. Slack time is optimized by merging inefficient processors, and the deadline is computed based on the finish time of the task heterogeneously. The containerization is used with Kubernetes as a package instead of Virtual Machines (VMs), which has advantages over the Docker Swarm. The initial result shows that the novel hybrid scheduling algorithm reduces approximately fifty percent of energy consumption balancing the scheduling task in parallel applications, but not in serial tasks.