Research On Tasks Cheduling Strategy In Cloud Computing Environment

Cloud computing technology provides users with more convenient and secure network services,users only need to use on demand and pay by volume.However,the scale of cloud platform cluster is huge,there is a great heterogeneity between cluster nodes,and the application requests of users are various.The rational division of resource types and the efficient scheduling of cloud computing tasks have become the key and difficult points in the field of cloud computing.In addition,there are obvious shortcomings in most traditional cloud computing task scheduling algorithms,so it is necessary to further study and improve the task scheduling algorithm on the basis of predecessors,so that it can be better applied to the field of cloud computing.Firstly,aiming at the deficiency of traditional scheduling algorithm in cloud computing platform,based on the standard differential evolutionary algorithm,Levy variation strategy is used to produce offspring,so that the fixed crossover factor and indentation factor in the algorithm adapt to the evolution of the population,and an improved differential evolutionary algorithm,LADE,is proposed.Compared with the standard differential evolutionary algorithm,the improved algorithm converges faster,the global optimization ability is stronger,the diversity of understanding is better maintained.Secondly,with the improved LADE algorithm as the optimization framework,combined with the characteristics of cloud computing task scheduling,taking into account the interests of users and the income of cloud service providers,this thesis analyzes and designs the task scheduling model based on LADE,and realizes the optimization algorithm of cloud computing task scheduling,which satisfies multi-Qo S constraints.Finally,the cloud task scheduling algorithm which improves the differential evolution algorithm is simulated on the Cloud Sim,and the experimental results are compared with other task scheduling algorithms,and the experimental results verify the validity and feasibility of the LADE task scheduling strategy proposed in this thesis.Effectively reduces the time it takes to complete tasks and the energy consumption of tasks,increases user satisfaction,and maximizes the revenue of cloud service providers.