Energy consumption control method of cloud data center for burst load

Abstract

The invention discloses an energy consumption control method of a cloud data center for a burst load, which optimizes a placement method of a virtual machine to reduce the energy consumption of the data center on the premise of ensuring normal operation of the data center when the data center is in the burst load and meeting the delay requirement of a user. Overall modeling is carried out in combination with a data center airflow organization mode under the condition of considering the sudden load, and the model is solved by utilizing an improved simulated annealing algorithm in combination with the load request quantity of the virtual machine, so that a virtual machine placement mode adaptive to the current data center airflow organization is obtained; meanwhile, due to the fact that the arriving task load has uncertainty under the sudden load condition, misjudgment is prone to being generated in partial virtual machine placement algorithms, and therefore the model and the method are used for optimizing the virtual machine placement strategy to adapt to the arriving task load of the current data center. Virtual machine placement is combined with the task amount arrived by the current data center, the operation performance of the data center is guaranteed, and a virtual machine placement scheme is optimized.

Description

technical field [0001] The invention relates to the technical field of data center resource scheduling optimization, in particular to a method for controlling energy consumption of a cloud data center in the face of sudden loads. Background technique [0002] With the increasing popularity of cloud computing, the size and business volume of data centers are gradually increasing. The service level agreement (SLA) between the service provider and the user specifies the guarantee of the task delay. The service provider must provide resources according to the user's needs, and meet the user's SLA while ensuring the completion of the task. This brings challenges in the following aspects. [0003] First of all, how to effectively handle a large number of sudden workloads that arise during the operation of the data center has become a difficult problem. When faced with a sudden load, service providers usually use methods such as increasing the number of servers or waiting for ta...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the above-mentioned defects in the prior art, and provide a method for controlling energy consumption of a cloud data center in the face of sudden loads. The sudden load of the center can improve the problems of unstable operation, high energy consumption, and frequent hotspots when the data center encounters sudden load. By modeling the arrangement of data center servers and the distribution of airflow organization, the The incoming task queue is split and reorganized, and the simulated annealing algorithm is used to calculate the best solution for virtual machine placement, so as to improve the stability and efficiency of server operation and reduce the possibility of hot spot failures

Images