Data center global energy consumption optimization-oriented airflow perception type virtual machine scheduling method

Abstract

The invention discloses a data center global energy consumption optimization-oriented air flow perception type virtual machine scheduling method, which comprises the following steps: firstly, modelingthe energy consumption of an air flow organization model, a data center server model and a network energy consumption model of a data center, and solving an overall model by using a simulated annealing algorithm in combination with the load request quantity of a virtual machine; and obtaining the virtual machine placement mode adaptive to the overall energy consumption of the current data center.Meanwhile, the utilization rate of the virtual machine is greatly changed in the execution process due to the uncertainty of task load during task execution, so that the constructed overall model andthe scheduling method are used for scheduling the virtual machine to reduce the operation energy consumption of the current data center. According to the method, virtual machine placement and scheduling are combined with the overall energy consumption of the current data center, on the basis of reducing the refrigeration energy consumption of the data center, the server and network energy consumption during the operation of the data center is further reduced, the generation of local hot spots is greatly reduced, and the operation reliability of the data center is improved.

Description

technical field [0001] The invention relates to the technical field of data center resource scheduling optimization, in particular to an airflow-aware virtual machine scheduling method oriented to data center global energy consumption optimization. Background technique [0002] Cloud computing realizes on-demand provision of various computing services to users through the virtual aggregation and sharing of a large number of computing resources, so it can meet the growing demand for big data processing. In order to further improve the cloud computing system's ability to manage and process big data, how to reasonably and efficiently schedule cloud computing resources to provide users with elastic computing services is the key to improving the performance of cloud computing systems. [0003] In the cloud computing environment, due to the huge amount of cloud computing resources, the amount of computing services undertaken by the virtual machines in the data center is also quite...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the above-mentioned defects in the prior art, provide an airflow-aware virtual machine scheduling method for data center global energy consumption optimization, and improve the problems of high energy consumption and frequent hotspots in today's data centers. Model data center servers, networks and airflow organization, use simulated annealing algorithm to calculate the best solution for virtual machine placement and scheduling, improve server operating efficiency, reduce global energy consumption of data centers and reduce the possibility of hot spot failures

Images