Research On Virtual Machine Scheduling Strategy Based On Energy Consumption Awareness And System Implementation In Cloud Computing Environment

As a kind of distributed computing,cloud computing provides customizable virtual computing resources to users through the network.Due to its flexible and dynamically scalable characteristics,cloud computing has developed rapidly in recent years.In order to meet the growing needs of users,the scale of data center is expanding,resulting in a large amount of energy consumption and carbon dioxide emissions.The data center mainly manages physical resources through virtualization technology.Therefore,how to reduce the energy consumption of the whole data center through virtual machine migration is one of the main issues in cloud computing research.Virtual machine migration can reduce energy consumption by integrating virtual machines in the data center on a few active physical machines and shutting down the rest of the physical machines.It can be divided into three processes:physical machine classification,virtual machine selection and virtual machine placement.This paper studies the selection and placement of virtual machines in the above migration process.The specific contents are as follows:(1)A resource-aware VM selection algorithm based on prediction is proposed for the problem of inefficient migration due to the use of fixed objective function in the VM selection process.The algorithm combines the current resource utilization and predicted resource utilization to determine the physical machine state,classifies the resources on physical machines into overloaded resources and non-overloaded resources,and dynamically defines the selection function according to the usage of different physical machines,so that it can achieve the purpose of reducing the load of physical machines with fewer migrations and improve the migration efficiency.The experiments show that the proposed algorithm can effectively reduce the data center energy consumption,the number of migrations and the SLA violation rate.(2)In view of the problem that the virtual machine placement algorithm is easy to lead to the imbalance of multi-dimensional resources on the target physical machine,resulting in the waste of resources,the concept and calculation formula of the fit degree between virtual machine and physical machine are defined,and a virtual machine placement scheme based on improved ant colony algorithm is proposed.Experiments show that the proposed algorithm can effectively reduce the number of active physical machines and reduce the energy consumption of data center.(3)Design and implement a cloud platform control system to manage the physical machines and virtual machines in the cloud platform.The system provides virtual machine services to users,and the main functions include physical machine management,virtual machine management,and cloud platform status monitoring.The usability of the system is verified by system testing.