Virtual Machine Resource Prediction Model Based on Deep Learning

With the development of big data and artificial intelligence, various industries have accumulated a large amount of data, and the demand for computing power is increasing. In order to meet the growth of computing demand, the scale of cloud data centers is gradually increasing, the energy consumption is gradually increasing, and the architecture is also becoming more and more complex. The main carrier of virtualization in cloud data center is virtual machine, so the study of live migration of virtual machine plays an important role in reducing energy consumption and guaranteeing SLA(data center Service level Agreement). Live migration of virtual machines includes overload detection of physical machines, selection of virtual machines to be migrated and placement of virtual machines. By accurately predicting the resource requirements of the virtual machine at the next time node, the virtual machine live migration algorithm can be designed, which is helpful to select the appropriate virtual machine for migration and place the virtual machine on the appropriate physical machine, so as to reduce the energy consumption of the cloud data center and ensure the SLA of the cloud data center. This paper proposed a hybrid model using filters based on deep learning time series prediction to predict the resources of VM. The accuracy is improved compared to other methods.