Multi-objective Cloud Computing Task Scheduling Considering Task Emergency

The emergence of COVID-19 has promoted the process of digital transformation of enterprises,and promoted the rapid development of cloud services such as cloud office,online education and so on.With the rapid growth of cloud computing task demand and the continuous increase of cloud computing resources,how to allocate resources scientifically and reasonably through task scheduling algorithm has become an important research problem in the field of cloud computing.Cloud computing is a business service model.In cloud task scheduling,we need to protect the diverse needs of users as much as possible and improve user satisfaction.This thesis proposes the maximum response time of the task to measure the urgency of the task,optimizes the average response timeout of the task to ensure the quality of service of users,and establishes a multi-objective cloud computing task scheduling model considering the urgency of the task.Genetic algorithm is used to solve the model,and an extended arrangement is designed for chromosome coding to optimize the task execution order.Aiming at the problems of premature convergence and low accuracy in solving the model of genetic algorithm,the parameters of population diversity and convergence degree are introduced to adaptively adjust the crossover and mutation probability of genetic algorithm,which improves the dynamic search ability of the algorithm.The simulation on cloudsim platform shows that the improved genetic algorithm has better performance in solving cloud computing task scheduling problems of different scales.Considering that different decision makers have different preferences for optimization objectives,this thesis uses NSGA-II to solve the multi-objective cloud computing task scheduling problem.Based on NSGA-II algorithm,a dynamic adaptive individual selection method is proposed to maintain the diversity and distribution of evolutionary population.The experimental results show that compared with other traditional algorithms,the improved NSGA-II algorithm has better convergence accuracy and diversity in multiobjective cloud computing task scheduling,and can effectively reduce the task completion time while considering the task urgency.