Research On Cloud Computing Resource Scheduling Based On Improved Genetic-Ant Colony Algorithm

With the continuous increase of network bandwidth,access to non-local data processing and computing services has become more frequent.In order to better store and process large-scale data,the concept of "cloud computing" came into being.Cloud computing is a new type of computer network technology and service mode.It has the characteristics of flexibility,pay-per-demand,etc,which can bring comprehensive and effective services to the users and bring huge profits to enterprises.It has a very good development prospect.However,cloud computing deals with resources and information by focusing on the cloud.Under such a large-scale processing approach,the strategy of resource scheduling is very challenging.If the scheduling strategy is unreasonable,it will inevitably cause time is long-term,high-cost and waste of resources and energy consumption increase a series of problems.Therefore,the establishment of a reasonable and effective scheduling strategy is a difficult problem to be solved.Currently,the resource scheduling algorithms in the cloud computing environments have transitioned from traditional algorithms to heuristic algorithms.For cloud scheduling algorithms,deeper exploration and application have been made on the basis of scientific achievements of other scholars.1.The current research status of cloud computing resource scheduling algorithms,genetic algorithms,and ant colony algorithms is analyzed.The initial search efficiency of the genetic algorithm is high,and the local optimum solution is easily generated in the later period.The initial search efficiency of ant colony algorithm is low.Because of the positive feedback characteristic such as easy to get the optimal solution.It is easy to get the optimal solution at later stage due to its positive feedback and other characteristics.The genetic algorithm and the ant colony algorithm are improvedrespectively and then the dynamic scheduling algorithm is combined.2.The resource scheduling and classical algorithms of cloud computing technology are discussed in detail.Combined with the principles of genetic algorithms and ant colony algorithms,they were separately improved,the genetic algorithm is introduced in time-load dual fitness function,using contemporary population average fitness value and best adapted to calculate the crossover,mutation rate of the adaptive value.In the ant colony algorithm,time-cost double functions are introduced to update the pheromone of the ant colony algorithm.The pheromone boundary value is defined by the maximum and minimum values in the maximum and minimum ant systems,and the transition probability is improved by the pseudo-random proportion rule.Increase its randomness and define the value of heuristic information through the load balancing function.3.Using dynamic fusion strategy to transform the top 10% of the optimal solution of genetic algorithm into the initial pheromone distribution of ant colony algorithm.CloudSim through cloud computing simulation tools for improved genetic-ant colony dynamic fusion algorithm,including three aspects of algorithm execution time,cost,and load imbalance.Comparing the running results with the traditional polling algorithm,the improved genetic algorithm and the improved ant colony algorithm,the effectiveness of the algorithm in the process of cloud computing resource scheduling is proved.