Research On Energy Optimization Method Of Cloud Computing System

High energy consumption is a major problem in the cloud computing system. Moreover, with the growing scale of cloud computing in recent years, the energy consumption is becoming more and more serious. The thesis focuses on the research about energy consumption optimization method for idle and extravagant energy consumption in order to reduce energy consumption.Firstly, traditional energy consumption optimization technology is summarized. And then the current situation of research on resource hibernation technology and resource scheduling algorithm applied in the field of cloud computing energy consumption optimization are analyzed.Secondly, considering the different states of resources and different types of tasks in the cloud computing systems, the MSEM (Multi-State Energy Model) of cloud computing is proposed in the thesis and is verified by experiment.In order to reduce the idle energy consumption, an adaptive idle time predictive strategy based on exponential average algorithm is proposed. The method imports the adaptive regulatory to adjust the influence of predicted idle time from historical idle time dynamically. Besides, the method imports the idea of segmented sliding window. The idle time within the sliding window is divided into three categories as long, medium and short which can reduce the bad influence of accidentally undulating. The experimental results show that the method has high rate of prediction accuracy, little system response and low energy consumption.In order to reduce the extravagant energy consumption, based on Min-Min algorithm and on the premise of meet the QoS, considering the energy consumption and load balance factor, a resource scheduling method Energy-Aware Min-Min algorithm is proposed. The simulation on CloudSim platform results show that Energy-Aware Min-Min algorithm achieves the purpose of energy saving.