Algorithm for Historical Periodic Load Oriented to Virtual Machines
FU Yinsheng
ZHAO Wenhao
CHEN Yuqi
Abstract：Deploying various services as virtual machines in data centers has become the mainstream method.To improve the resource utilization of physical machines in data centers,hot migration technology has been developed for online scheduling of virtual machines between physical machines.However,frequent hot migration consumes additional resources and can cause temporary downtime of virtual machines,affecting service stability.Ensuring service stability while maintaining high resource utilization has become a critical issue that needs urgent resolution.Therefore,this paper proposes an algorithm based on historical periodic load prediction(BOHPLF),which can separate virtual machines with predictable periodic loads.The author uses real datasets from the Azure cloud for simulation experiments,and the results show that 60.96％of virtual machines have stable and predictable periodic loads.
Keywords：cloud computingperiodic load forecastingresource utilization rateenergystability of service
Publication Date：2024-07-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 55-59 )