Design And Implementation Of Workflow Task Allocation Based On Ant Colony Algorithm

Cloud computing is a service that provides users with computing resources over the Internet and handles the demands of users.In recent years,with the rapid development of cloud data centers,there are more and more users of cloud computing services,which has led to a significant increase in demand for cloud data center services.At the same time,many complex tasks begin to gradually execute in a cloud environment in the form of workflow.Therefore,for cloud service providers,how to effectively schedule the execution order of workflows so as to minimize the completion time of workflows has become a very important research issue at present.Ant colony algorithm is a heuristic optimization algorithm inspired by the phenomenon of finding the shortest path in the process of ant foraging.Because of its ability to find better solutions and the positive feedback and robustness of the information,ant colony algorithm has a strong advantage in solving the task assignment NP-hard problem.In view of this,this paper designs and implements a workflow task allocation mechanism based on ant colony algorithm,which minimizes the completion time of the workflow by optimizing the resource allocation of the workflow.The main work and contribution of this article are as follows:First of all,the task scheduling problem of the cloud workflow management system is analyzed and studied.The DAG model is used to describe the task execution flow of the workflow,and the optimization model of the task scheduling problem of the cloud workflow is built,with the objective of minimizing the total completion time of the workflow.Secondly,an ant colony optimization algorithm is proposed to solve the task scheduling problem of cloud workflow system.The goal of the algorithm is to find a mapping between the virtual machine and tasks,so as to minimize the completion time of the workflow.Compared with the existing cloud workflow task scheduling algorithm,the proposed algorithm combines an ant system algorithm with a local search algorithm to more effectively and globally search the solution space,thus a higher quality solution can be found.Finally,the effectiveness of the proposed ant colony algorithm is verified.First,the convergence of genetic algorithm and ant colony algorithm is compared with Mongate workflow application.Then,in the five types of workflow applications,ant colony algorithm,genetic algorithm,random algorithm and HEFT algorithm are run respectively.By comparison,it is found that the completion time of ant colony algorithm in the five types of workflow applications is less than the other three algorithms.Ultimately,we conclude that ant colony algorithm can improve the task allocation efficiency of workflow.