Research On Distributed Intelligent Process Scheduling

Cloud computing R & D is a great change in the information age,which has driven the development of new information industry.In cloud computing,cloud vendors manage a large number of computing resources,storage devices and so on,and then provide them to users in the form of network services.The emergence of cloud computing realizes the sharing of computing resources.However,a large number of resources are unified,which makes its architecture more complex.With the development of the network,more and more cloud users,and the amount of tasks processed by the system doubled.Therefore,it is more important to allocate and schedule the tasks reasonably.It is a big challenge in the field of cloud computing to meet the needs of user services and rationalize the use of system resources.Through the research of cloud computing,cloud task scheduling technology and theory,and aiming at the problem of data constraint task scheduling in big data environment,the research work of this paper mainly includes the following two points:1.In order to solve the multi-objective optimal scheduling problem of workflow with data constraints,a data aware scheduling algorithm based on business logic and considering data constraints is proposed.On the basis of ant colony algorithm,this algorithm judges the amount of data needed by each business and the data constraints between tasks,perceives the data flow between each business,and then uses the divide and conquer strategy to search the optimal scheduling.In this paper,the improved algorithm is tested and evaluated in Cloud Sim simulation platform.The experimental results show that the algorithm is superior to the existing solution methods in efficiency and cost-effectiveness.Finally,the scheduling management module is implemented on the distributed business process platform,which shows the effect of the proposed scheduling algorithm in practical application.2.Based on Padres system,an intelligent process management and control middleware system platform is developed,on which a distributed process management engine is implemented to meet the needs of cloud service resource integration in cloud computing and big data environment.Then the scheduling algorithm model and process management platform are combined to realize a cloud computing process management system by using front and back technology.