Research On Task Scheduling Based On Particle Swarm And Ant Colony Algorithm For Cloud Computing

In view of the huge amount of Internet application data and its explosive growth, with the massive data storage and processing, cloud computing technology comes into being. Cloud computing task scheduling is a key technology of cloud computing. A good task scheduling algorithm, not only can help us to build a stable, robust and energy-efficient cloud computing environment, but also can improve the user satisfaction with the use of cloud computing services. In this paper, we mainly study the task scheduling problem based on task completion time optimization, and propose several task scheduling algorithms, which have very important theoretical significance and practical value. The creative work of this paper is embodied in the following several aspects.For a single scheduling algorithm is difficult to adapt to different types of tasks in the cloud computing environment, a multi-level queue scheduling strategy is proposed. This strategy puts different types of tasks into different queues according to the task priority order, which can effectively meet the requirements of cloud computing task scheduling diversity. In view of the strategy, a simple and efficient task scheduling algorithm is designed, which based on resource and task matching, not only optimizes the task completion time, but also balances the resource load pressure. The experimental data shows that the combination of the algorithm and the multi-level queue scheduling strategy has a remarkable experimental effect in the complex cloud environment.Aiming at the problems that traditional scheduling algorithms are difficult to achieve multi-objective optimization, the Extremum Disturbed Correlative Particle Swarm Optimization algorithm is proposed. The algorithm uses the Copular function to establish the correlation between random factors, which resolves the demerit that the Particle Swarm Optimization algorithm lacks of the global optimization ability because of not considering the function of the random factors in the optimization process. And adds the extremum disturbance operator to resolve the problem of the low convergence accuracy of the Particle Swarm Optimization algorithm. The simulation experiments show that the proposed algorithm is better than the standard Particle Swarm Optimization algorithm and the traditional FIFO scheduling algorithm, which is an effective task scheduling algorithm.At last, by analyzing the advantages and disadvantages of Particle Swarm Optimization and Ant Colony Optimization algorithm, a task scheduling algorithm based on fusion of the advantages of the two algorithms is proposed. The algorithm divides the total iterative process into two stages: the first stage uses Particle Swarm Optimization algorithm to achieve the fast convergence of cloud computing task scheduling. The second stage uses Ant Colony Optimization algorithm to improve the ability to optimize the resources in the process of cloud computing task scheduling. The experimental results show that the proposed algorithm is better than the each algorithm which is used alone.