The Construction Of Antenna Simulation Cloud Platform Based On OPENSTACK And Its Resource Scheduling Algorithm Optimization

With the popularization and development of smart antenna and antenna simulation technology,the demand of research institutions,universities and small and medium for computer hardware and software infrastructure also has a rapid growth,frequent renewal of equipment and relatively independent software and hardware operation mode,leads the cost of antenna simulation to increase year by year.With the rise of cloud computing and virtualization technology,antenna simulation and cloud computing technology can be combined to break the bottleneck of the development of simulation technology.The simulation of antenna cloud platform of university or small and medium enterprises,not only can make the rational use of computer software and hardware resources,and can greatly facilitate the practice of scientific research personnel,improve the efficiency of scientific research.This paper first introduces several mainstream IaaS cloud platforms,and selects the OpenStack which is widely accepted by industry to construct the antenna simulation cloud platform,to create simulation environment file template and cloud hosting type,to provide users simulation environment on demand and the resource of computer hardware;Next,this paper conducts an in-depth 'study of the overall architecture and core components of OpenStack,and combined with the OpenStack source code,the initial virtual machine allocation module,dynamic virtual machine migration module are analyzed in detail,and expounds its working principle,points out the shortcomings of them;finally,aiming at the defects of OpenStack native resource scheduling strategy,the paper presents a multi-objective improved ant colony algorithm and multi-objective optimization algorithm for dynamic migration of virtual machines.Using the improved multi-objective ant colony algorithm to optimize virtual machine placement strategy,the algorithm is the improvement of ant colony algorithm,which obtains the optimal solution through continuously updating pheromone,and then finds the best position for the virtual machine newly created.The simulation results show that the proposed algorithm can not only ensure favorable service performance,but also can reduce the load and power consumption of resources,and insure the data center to achieve a robust running state.The dynamic migration strategy of OpenStack virtual machine is optimized by using the multi-objective optimization algorithm.The design of two points on the whole time delay method and prediction method to determine the migration occasion,using the multi-objective optimization algorithm to select the appropriate physical host,and the probability selection algorithm is used to avoid the bunching effects of virtual machines.The simulation results of CloudSim show that the set of virtual machine resource dynamic scheduling method can be qualified for real-time scheduling of resources,and can optimize the performance of data center.