Energy Optimization Oriented Three-Way Clustering Algorithm for Cloud Tasks
Energy Optimization Oriented Three-Way Clustering Algorithm for Cloud Tasks

Cloud computing has developed as an important information technology paradigm which can provide on-demand services. Meanwhile, its energy consumption problem has attracted a growing attention both from academic and industrial communities. In this paper, from the perspective of cloud tasks, the relationship between cloud tasks and cloud platform energy consumption is established and analyzed on the basis of the multidimensional attributes of cloud tasks. Furthermore, a three-way clustering algorithm of cloud tasks is proposed for saving energy. In the algorithm,first,the cloud tasks are classified into three categories according to the content properties of the cloud tasks and resources respectively. Next, cloud tasks and cloud resources are clustered according to their computation characteristics (e.g. computation-intensive, data-intensive). Subsequently, greedy scheduling is performed. The simulation results show that the proposed algorithm can significantly reduce the energy cost and improve resources utilization, compared with the general greedy scheduling algorithm.
Cloud computing has developed as an important information technology paradigm which can provide on-demand services. Meanwhile, its energy consumption problem has attracted a growing attention both from academic and industrial communities. In this paper, from the perspective of cloud tasks, the relationship between cloud tasks and cloud platform energy consumption is established and analyzed on the basis of the multidimensional attributes of cloud tasks. Furthermore, a three-way clustering algorithm of cloud tasks is proposed for saving energy. In the algorithm,first,the cloud tasks are classified into three categories according to the content properties of the cloud tasks and resources respectively. Next, cloud tasks and cloud resources are clustered according to their computation characteristics (e.g. computation-intensive, data-intensive). Subsequently, greedy scheduling is performed. The simulation results show that the proposed algorithm can significantly reduce the energy cost and improve resources utilization, compared with the general greedy scheduling algorithm.