Energy Efficient Container Consolidation Method In Cloud Environment Based On Heuristic Algorithm

Since its birth,cloud computing has been rapidly developed and widely used due to its high availability,scalability,and cost-effectiveness.The rapid development of cloud computing makes huge data centers require huge amounts of power supply to maintain operation.As the scale of cloud service providers expands,the power consumption of infrastructure is likely to become the biggest bottleneck restricting the development of data centers.Whether it is from the perspective of improving the economic benefits of cloud service providers,or from the perspective of protecting the environment and reducing the carbon dioxide emissions that cause global warming,increasing the energy utilization rate of data centers is an unavoidable topic.Based on the popularity of cloud computing,academia and industry have conducted extensive and extensive research on the energy-saving methods of virtual machine-physical machine architecture.Cloud container is a new service architecture after infrastructure as a service,platform as a service and software as a service.Since its birth in 2015,it has occupied 13% of the cloud service market within a year,but the energy conservation research on cloud container architecture is second to none.countable.Therefore,this paper discusses the energy saving of the cloud container environment,and proposes a heuristic dynamic container integration method to maximize the resource utilization rate of the data center.The principle of the container integration method to reduce energy consumption is to concentrate the load of the data center on as few servers as possible and shut down or sleep idle hosts.However,the concentration of the load on a small number of servers will increase the risk of server overload,which in turn will cause the service level to decline.In order to avoid the occurrence of overload while integrating containers,this paper designed a prediction algorithm to determine the risk of host overload,so that the host can be migrated before the host actually enters the overload state.And on the basis of this method,the quality of service has been further improved.The improved algorithm selects the destination host for the container through an adaptive resource reservation mechanism when the container is migrated.This mechanism avoids the possibility of service quality degradation by reserving idle resources on the host to avoid repeated overloading of the host due to load fluctuations after the execution of the migration strategy.In order to prove the effectiveness of the design method,the paper conducted an experimental evaluation based on the workload record of the real environment.Experimental results show that the two heuristic container integration algorithms designed significantly reduce the energy consumption of the data center,and the improved algorithm can reduce the energy consumption of the data center while avoiding the service level reduction as much as possible.Cloud container resource scheduling itself is an NP-hard problem.The advantage of the heuristic algorithm is that it can find a feasible solution for the NP-hard problem within an acceptable time cost.This paper realizes that optimal energy efficiency can be achieved in similar algorithms.The two optimization strategies and optimal service quality can be applied to cloud service environments with different requirements.