Research On Virtual Machine Placement Method Of Cloud Platform For Multi-objective Optimization

In recent years,cloud computing has developed rapidly.As an efficient technical model,cloud computing has attracted extensive attention of many enterprises at home and abroad because of its many advantages.Among them,many Internet leading enterprises such as Google,Microsoft and Alibaba have invested a lot of costs in the cloud computing industry and established many data centers around the world.With the explosive growth of the number of machines in the data center,the problem is also very serious,that is,the high energy consumption and inefficient utilization of resources in the data center.The key to solve this kind of problem is to reasonably place virtual machines to make full use of resources in the data center.Although there have been abundant research results at home and abroad,there is still room for further optimization.Therefore,this paper proposes two different virtual machine placement strategies to optimize the energy consumption,resource utilization,resource fragmentation and robustness,so as to make the resource use of the data center more efficient and greatly reduce the energy consumption.The main work of this paper consists of the following two parts.(1)Aiming at the problem of resource waste and inefficient utilization caused by the accumulation of a large number of resource fragments in the process of virtual machine placement,this paper proposes a virtual machine placement method MALO(Modified Antlion Optimization)based on the improved antlion algorithm.In addition to the optimization of energy consumption and utilization rate of resources,the optimization method is also considered in the process of resource placement.Reducing the number of resource fragments is conducive to reducing resource waste and making full use of resources.Experiments on the proposed MALO algorithm and several other algorithms are carried out on the Python simulation platform.The results show that the proposed MALO algorithm can not only effectively reduce energy consumption and improve resource utilization,but also reduce the number of resource fragments.(2)In order to further optimize the overall robustness of cloud data center on the basis of reducing energy consumption and resource fragments,this paper proposes a virtual machine placement strategy FFA-VMP(Firefly Algorithm-Virtual Machine Placement)based on improved firefly algorithm.By changing the step size factor of the traditional firefly algorithm to adaptive change,and grouping and moving all firefly individuals,the placement algorithm can effectively avoid falling into local optimization and better adapt to the coding of virtual machine placement in cloud environment.In order to verify the effectiveness of the placement algorithm,the proposed algorithm and several other algorithms are tested on the Python simulation platform.The results show that the proposed FFA-VMP algorithm not only ensures low energy consumption and low resource fragmentation,but also improves its robustness,so as to ensure good quality of service.