Research On Scientific Workflow Scheduling Method Based On Multi-object In Cloud Computing Environment

With the advent of the intelligent era,more and more experimental mod els have turned to cloud computing and data storage platforms.A large number of workflows have accumulated in resources,which has caused scientific workflow scheduling to become an urgent problem to be solved.Existing scheduling research objectives are relatively simple,ignoring the partial order relationship between tasks and the differences in computing resources,resulting in scheduling strategies that cannot meet diverse application scenarios.Therefore,in this paper,we will study the multi-objective scheduling of scientific workflow in different cloud environments for the above problems.The specific research contents are as follows:Aiming at the problems of minimizing workflow completion time,scheduling fairness and cost optimization in a public cloud environment,a task scheduling algorithm based on partial order relationship is proposed.The algorithm generates scheduling plans based on the partial order relationship between tasks.Consider the global impact of single task node scheduling to a void unfair scheduling that will prevent subsequent tasks from being completed within the deadline.At the same time,the completion time of the task node is minimized and the cost is reduced.Aiming at the problem of user's dynamic submission of workflow,based on the task scheduling algorithm based on partial order relationship,a dynamic programming scheduling execution algorithm is proposed.Perform dynamic execution according to the scheduling plan,and adjust in real time according to the scheduling information.Combine new tasks submitted by users with old tasks to generate new scheduling parameters.To avoid prior scheduling of old tasks and lead to large scheduling time of new tasks,to ensure the fairness of the scheduling process.Aiming at the problems of heterogeneous resource distribution and large differences in performance indicators in a hybrid cloud environment.Based on the improved task scheduling algorithm based on partial order relationship,a resource-driven optimal scheduling algorithm for heterogeneous hybrid cloud is proposed.Improve the task priority calculation strategy and resource traversal ideas.Use improved strategies to pre-process tasks and resources.Scheduling through resource-driven heuristic algorithms.Driven by idle resources,trigger scheduling mechanism.Reduced idle waiting time of resources and improved resource utilization.Finally,through the Cloudsim simulation experiment platform,the simulation experiments of the proposed algorithm are compared.The dynamic programming and scheduling algorithm based on partial order relationship evaluated the experimental results from multiple angles.Verifies that the algorithm performs better than existing algorithms in terms of scheduling length,running time,fairness and cost.The resource-driven optimization scheduling algorithm of heterogeneous hybrid cloud is found through experiment comparison.Not only has time and cost performance been improved,but resource utilization performance is significantly better than existing algorithms.Optimized multi-objective scheduling of scientific workflow in cloud environment.