Research On Computing Resource Scheduling Algorithm In Cloud Environment

The scheduling strategy on the cloud computing platform is very important to the operation efficiency of the cloud computing system.Appropriate scheduling strategies can effectively reduce the completion time of cloud tasks,reduce system operating costs,and ensure load balance among computing nodes.In order to optimize the operation efficiency of the static and dynamic resource scheduling process,this paper establishes a static resource scheduling model and a dynamic resource scheduling model based on the cloud task allocation model proposed by Google and the cloud computing simulation platform developed by the University of Melbourne.According to the characteristics of the two models,two different hybrid optimization algorithms are proposed.Finally,the performance of the proposed two hybrid algorithms in different task scheduling models is verified by simulation scheduling.Aiming at the static resource scheduling model,a hybrid algorithm of improved genetic algorithm and beetle search algorithm is proposed.The hybrid algorithm adjusts the crossover method according to the iterative stage,which not only speeds up the convergence speed of the algorithm in the early stage,but also improves the overall optimization ability in the later stage of the algorithm;dynamically adjusts the mutation probability according to the fitness of the individual,and improves the optimization ability of the algorithm for inferior solutions;adopts the elite solution retention strategy,Protect the high-quality solution from being destroyed;introduce an improved beetle search algorithm to improve the local optimization ability of the genetic algorithm.The simulation results show that compared with the comparative optimization algorithm,the proposed improved genetic beetle hybrid algorithm can effectively reduce the task completion time,indicating that the hybrid algorithm proposed in this paper is an efficient and fast optimization algorithm in static resource scheduling.Further considering the problem of computing resource changes due to various uncontrollable factors in the scheduling process,a multi-objective dynamic resource scheduling algorithm based on a mixture of improved artificial immune algorithm and simulated annealing algorithm is proposed.The proposed dynamic scheduling algorithm first improves the immune operation,changes the structure of the antibody gene in a large range,ensures the rapid convergence of the algorithm,and improves the real-time performance of the algorithm;adds a variety of neighborhood search operations to improve the local search ability of the algorithm due to rapid convergence.At the same time,in order to solve the change of the feasible solution space caused by the dynamic change of resources,the simulated annealing algorithm is further mixed to enhance the global search ability of the algorithm in the feasible solution space,and effectively improve the real-time optimization ability of the algorithm.The final simulation results show that the proposed dynamic resource scheduling algorithm based on the combination of improved immune algorithm and simulated annealing algorithm has the best optimization effect in reducing task completion time,reducing task execution cost and ensuring computing node load balance when faced with changes in computing resources.Compared with other algorithms,it shows that the improved artificial immune and simulated annealing hybrid algorithm proposed in this paper is an efficient multi-objective dynamic resource scheduling algorithm.