Research On Grid Task Scheduling Method Based On Dependency Task

Grid computing is a new research topic in the field of computer science,which is a very high academic and application value.How to use grid resources effectively is the core of research.Task scheduling is one of the core services in grid computing.In this paper,based on the research of grid computing theory,the paper has carried on the thorough research to the different task scheduling algorithms.A priority based task scheduling algorithm is proposed,which is based on the task model and resource model.There is a dependent relationship between tasks,the priority method of the task is designed.This method gives a priority to each task.Tasks filtered by the priority number.According to the proportion of the length and length of the task set,combined with the Min-Min and Max-Min algorithm for task scheduling,to minimize the completion time scheduling objectives.An adaptive fault tolerant scheduling algorithm based on dependent task is proposed.The algorithm takes into account the task of fault tolerance,and ensure that the task can be performed correctly.After studying the basic method of fault tolerance,the algorithm uses adaptive fault tolerance algorithm.In order to reduce the cost of the backup,the backup level is introduced in order to reduce the cost.The backup tasks can be divided into active and passive parts,which can effectively reduce the time and cost of the system.These two algorithms are compared with the classical task scheduling algorithm for simulation experiments.The results verify the correctness and superiority of the proposed algorithm used in the grid environment.On the basis of in-depth study of grid computing,according to the requirements of the information grid technology research project,a prototype of grid computing is designed and made.The software provides a variety of functions for the user to carry out the research work of the grid computing.