Research On Virtual Machine Placement Problem Based On Intelligent Optimization Algorithm

With the increasing scale of the cloud computing market,cloud data center energy consumption is also increasing dramatically year by year.The increase in energy consumption not only makes the operation cost too high,but also leads to a rapid increase in global greenhouse gas emissions,which has a great negative impact on climate and environment,so it is crucial to reduce data center energy consumption.Proper placement of virtual machines is an important way to reduce data center energy consumption.Some results have been achieved in the area of virtual machine placement,however,most of the studies have focused on optimizing energy consumption and resource utilization,while less consideration has been given to the network communication overhead and ensuring the quality of service for users generated by virtual machine placement.Therefore,in this paper,we propose two improved intelligent optimization algorithms for virtual machine placement considering the above objectives when studying the virtual machine placement problem.The main work of this paper is as follows:(1)A virtual machine placement method(WOAGS)based on the improved whale optimization algorithm is proposed with the optimization objectives of reducing energy consumption and communication cost.The basic whale optimization algorithm is improved for better application in virtual machine placement in order to address the shortcomings of the basic whale optimization algorithm such as low solution accuracy,slow convergence speed and easy to fall into local optimality when solving complex optimization problems.By optimizing the convergence factor and step size factor,the algorithm can still have the probability to perform global search and jump out of local optimum in the late iteration;introducing Levy flight strategy to expand the search range and increase the population diversity;incorporating golden sine algorithm to effectively narrow the search range,speed up the convergence speed and improve the convergence accuracy.The performance test of the improved algorithm was conducted to verify the superiority of its search effect.Finally,the improved whale optimization algorithm WOAGS is used to solve the virtual machine placement problem,and the placement strategy based on traffic tightness is adopted in the initialization,and the experimental comparison with several other placement algorithms shows the superiority of the performance of the improved algorithm in optimizing the placement target.(2)A virtual machine placement method(ISSA)based on an improved sparrow search algorithm is proposed with the optimization objectives of reducing energy consumption,resource waste and SLA violation rate.Improvements are made to address the problems of the sparrow search algorithm applied to virtual machine placement such as easy to fall into local optimum and insufficient population diversity.The crisscross strategy is introduced for optimization after the explorer position update to improve the defect that the algorithm converges to the local optimum solution too early in the early iteration;the Levy flight mechanism is introduced into the follower position update so that it can expand the search range and increase the diversity of the population;after the population information is updated,a dynamically adjusted lens imaging reverse learning strategy is used to solve the reverse solution for the population individuals to enhance the population The ability to escape from the local optimum.The improved algorithm is tested for performance using test functions to demonstrate its superior optimality finding capability.Finally,the improved ISSA algorithm is applied to the virtual machine placement problem in this chapter,using a fuzzy evaluation-based approach to make adjustments to the fitness function.By conducting simulation experiments for comparison,it is shown that the improved algorithm is relatively more effective in optimizing the placement target.