Research On Virtual Machine Deployment Technology In Cloud Computing

As an important reform in the field of computer industry recently, cloud computing has developed rapidly and provided almost unlimited resources such as virtual computation, storage and network for users. With the constant development of cloud computing technology currently, how to deploy virtual machines on physical server efficiently has become an important issue. How to reduce energy consumption under cloud computing environment is always a research emphasis of scholars.Proposed algorithm:This paper proposes a multi-population ant colony algorithm. Algorithm strategies in dynamic ant, random perturbations ant colony, max-min ant system are adapted in this paper to optimize the proposed algorithm. By introducing adaptive optimization strategy in case of local circling and random disturbance factor to increase the possibility that each path being selected, the obtained solutions are supposed to be objective and diverse.This paper first use the proposed multi-population ant colony algorithm to solve the deployment problem of virtual machine. With resource wastage and energy consumption as optimization objects. The optimal solution for each population is sought independently and their information communication is realized through information entropy.The simulation results show that this algorithm has better performance than single-population ant colony algorithm and can reduce resource wastage and energy consumption effectively on high-demand virtual machine deployment. The optimization effect obtained by multi-population ant colony algorithm is more obvious and superior compared to single-population ant colony algorithm. This indicates that the coordination and cooperation among multi-population ant colonies through information entropy can solve the convergence problem of the algorithm better and meanwhile expand the search space and improve the diversity of solutions searched.