Research On Resource Allocation Strategy In Cloud Computing Environment

Due to the rapid development of information technology in various industries and the continuous renovation of Internet technology,people's working methods have undergone great changes.People start to complete the work with computers through Internet,it lead to soar the amount of data in the Internet,at the same time “the age of big data” begins to come.The normal way of work has been unable to match the changing needs of the task,“cloud computing” emerge in this case.Cloud computing is an innovative way of working,it combines high performance computing with network technology,and provides technical support for business computing of large enterprises.In the cloud computing environment,the specific distribution of computing resources is very complicated and the demand for resources will change dynamically.If underestimation of demand for client will lead to resource scale is not large enough to complete tasks,but overestimation of demand for client will also make many resources idle.The suitable resource allocation algorithm in cloud computing is the key to solve the above problems.After analyzing the disadvantages of ant colony in dealing with resource allocation problems,an improved method was provided: including the improvement of transition probability formula and the pheromone volatility coefficient,in order to improve the ability of global searching ?avoid the early converging and stagnation.Adding the load balance adjustment factor in the updating methods of local pheromone,in order to achieve the goal of uniform distribution of computer resources.Considering the blindness of ant colony algorithm in the initial stage of search,it is necessary to combine with genetic algorithm with ant colony algorithm.Because of the genetic algorithm has a fast convergence rate in the early stage of the algorithm,so it could be used first to obtain an optimum solution,which would then be converted to be the initial pheromone of ant colony algorithm so that an optimized ant colony algorithm could be used for search.By expand the cloud computing simulation platform Cloudsim,the paper finish contrast experiment between different algorithm.The result show that iterations,time cost,power consumption and the load balance of improved algorithm are better than those of PSO-GA ? PSO-ACO and GA-ACO.