Research And Realization Of Task Scheduling Algorithm Based On Machine Learning In Secur IVCE Environment

The rapid development of the Internet has brought a large number of idle or unreasonable use of network resources.This situation puts forward higher requirements on the task scheduling algorithm for grid computing,cloud computing.At the same time,the network has brought potential threat to the security of the task scheduling.How to achieve the bestmatching of resources and tasks and ensure the security of transmission data through the various links of the scheduling,so as to provide a more reasonable,efficient and safe service to users with the idle network resources,which is a key issue in the research of computing platform provider.This paper is inspired by the core idea that need to aggregate in the virtual computing platform(iVCE),and it is a method to find a better way to solve the matching problem from other side of the task scheduling----task.In this paper,firstly,I studied the existing task scheduling strategy,and found that task of mining and research is very little compared to the resource in the task scheduling algorithm.In today's Internet era,a variety of new systems,new applications,new tasks appear,so it is necessary to study the potential characteristics of the task.This paper is trying to find the method of analyzing the "need" of need to aggregate.It is found that the important branch of the machine learning----clustering algorithm has the ability of data mining.After the study of many kinds of clustering algorithms,the fuzzy c-means clustering algorithm is selected as the basis of the research of the task,and the stability of the algorithm is improved by the initial clustering center of FCM algorithm.Then,based on the results of task clustering,a new task scheduling algorithm based on machine learning task clustering is proposed.Through the simulation experiments,it is verified that the task scheduling algorithm based on the mining tasks' inherent characteristics has certain effect on reducing the average execution time and improving the success rate of task execution on the premise of maintaining high resource utilization.With the development of the network,the security problem is also a big problem.Because the task scheduling in every link of the information with the user's privacy and directly related to the security of the system,so it is necessary to carry on the encryption processing parameters in the task scheduling to ensure the security of all aspects.In this paper,we study the related technologies of asymmetric encryption and symmetric encryption,and sum up their advantages and disadvantages respectively.In this paper,we choose the method of encryption and decryption combined with AES and RSA.With the characteristics of the communication content and architecture of the task scheduling system,the key generation and storage scheme of data task scheduling system are designed.According to the above discussion this paper design and implement a task scheduling system.And introducing every module function and design ideas of the task scheduling system.This paper also introduce including interface,database tables,etc..of every module.And the system was tested in two aspects,including task scheduling and communication encryption.Test results show that the system has a certain advantage in operation stability and success rate of the scheduling tasks.