Predicting Virtual Machine Resource Consumption Based on Optimized Grey Model

2019 International Conference on Advanced Electronic Materials, Computers and Materials Engineering

无线电电子学;计算机科学;材料科学

Aiwu, Shi^1 ; Di, Gao^1 ; Kai, He^1

Cloud Computing and Big Data Research Center, Wuhan Textile University, Wuhan

430200, China^1

关键词: Cloud environments;    GM (1 , 1) model;    Machine resources;    Physical resources;    Prediction model;    Resource consumption;    Sorting outs;    Utilization rates;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/563/5/052002/pdf
DOI  :  10.1088/1757-899X/563/5/052002

来源: IOP

【 摘 要 】

In order to relatively change the computing specifications of virtual machines (VMs for short) for the changing resource consumption of VMs, and reduce the impact of the changes, this paper proposes the optimized grey GM(1, 1) model to predict the resource consumption of VMs. Because VM resources may fluctuate greatly in a short time, the initial data are smoothed first, and then modeled, the model edge value is optimized according to the characteristics of cloud environment. Experiments show that the accuracy and stability of the prediction model can be improved by sorting out the initial data and optimizing the model. The prediction value is helpful to design the allocation scheme of VM resources, improve the utilization rate of physical resources in cloud environment.