Research on task scheduling of PDTs virtual machine in cloud computing environment

Research on task scheduling of PDTs virtual machine in cloud computing environment

Zhang Jie, Nanjing Information Engineering University

On the basis of analyzing the target, model, process and characteristics of cloud computing task scheduling, this paper mainly makes the following work: (1) introduces the definition, service type, architecture and characteristics of cloud computing. At the same time, the model, target and QoS requirements of task scheduling in cloud computing environment are analyzed, and the process and characteristics of cloud computing task scheduling are described. (2) By studying the independent task scheduling model and the workflow task scheduling model under the existing cloud computing environment, a new task scheduling model--pdts (Partial Dependent tasks, partially associated task) is proposed. It is an innovative way to analyze and deal with the user's multi form task in cloud computing environment. The model mainly considers the user's requirement of "deadline time" for submitting tasks, and minimizes the user's overhead. (3) Based on the study of Ant colony algorithm, an improved ant colony algorithm is proposed to solve the PDTs task scheduling model. The algorithm divides the ant colony into several subgroups, and selects a better pheromone updating method for each subgroup, and makes local optimization to obtain the whole optimal solution. (4) Based on the study of particle swarm optimization algorithm, an improved PSO algorithm is proposed to solve the PDTs task scheduling model. The algorithm also uses subgroup partitioning method to obtain local optimal solution and global optimal solutions by adjusting the velocity and displacement of the particles in the subgroup. (5) on the basis of deep research on Cloudsim open source cloud simulation platform, the improved ant colony algorithm and improved particle swarm optimization algorithm are used to simulate the cloudsim.

Research on task scheduling of PDTs virtual machine in cloud computing environment