Task Scheduling Optimization Strategy In Cloud Computing

Task scheduling is the core technology of cloud computing,task scheduling is one of the most important part of the task scheduling in the process of cloud computing,therefore,the optimization of task scheduling mechanism is an important method to strengthen the comprehensive performance of cloud computing.In order to improve the service performance of cloud computing more effectively,many scholars have studied the problem of task scheduling in cloud computing.Ant colony algorithm(ACO)is a global optimization algorithm,in cloud environment,and it has the characteristics of distribution,randomness and feedback.the problems of task scheduling mechanism can be effectively processed by the characteristics of ant colony algorithm.Ant colony algorithm is applied to solve the task scheduling problem,firstly,the pairing probability is calculated,the ant distributes the task according to the pairing probability by calculating the pairing probability of the task and the virtual machine,and the target solution is obtained when the algorithm is convergent.Because of the random selection method and the feedback mechanism,the convergence rate will be slow and precocious.Aiming at the problem of ant colony algorithm in the application of task scheduling,the improvement of ant colony algorithm is based on completion time of task and load balancing in this paper.The main work is as follows:(1)This paper analyzes the characteristics of task scheduling,and the principle of the scheduling algorithm is analyzed,it summarizes and analyzes the reasons for the inefficient computation of existing algorithms,and an improved method is proposed to the existing problems.(2)In order to improve the computational efficiency and the quality of optimal solution,ant colony algorithm is improved by combining the task scheduling mechanism in this paper,the pheromone updating rules of ant colony algorithm are improved by weighting,comprehensive performance is improved by dynamically updating the volatilization coefficient.In the process of updating local pheromone,the load weight coefficient of the virtual machine is introduced to make the assignment of the task reasonably.(3)The improved algorithm is simulated by CloudSim,and other algorithms are simulated in the same environment.The results of simulation experiments are compared with different algorithms,and the results are analyzed and studied.The experimental results show that the task scheduling strategy based on the improved algorithm has a reasonable allocation of tasks,the convergence speed of the algorithm and the total execution time are optimized.Finally,the results are analyzed and summarized according to the test results.