Research On The Scheduling Of Fog Computing Resources Based On Genetic Algorithm

Since the 21 st century,with the increase in the number of Internet of Things users,the amount of data transmission has increased rapidly,causing the cloud server to be overburdened.Traditional cloud computing can no longer meet the needs of rapid development of the Internet of Things.As an emerging computing model,fog computing provides a new way to reduce the pressure on cloud servers.The fog nodes deployed at the edge of the network can share the work of cloud computing,and handle some simple tasks in the terminal,but the rational use of the resources of the fog node is still a difficult point and focus.The problem of fog computing resource management and scheduling is the key to affect the performance of fog computing services.especially when large-scale service requests occur.If the resource scheduling problem is not effectively solved,it will increase service delay,reduce resource utilization and customer satisfaction.degree.Therefore,this paper has carried out in-depth research on the problem of fog computing resource management and scheduling,improved the existing fog computing framework,and proposed a proxy-based cloud fog joint framework.In the proposed framework,the fog computing resource management and scheduling strategy is established,and a reputation model for evaluating the credibility of fog nodes is established.The reputation model is established by verifying the historical status of the service provided by the fog nodes.By selecting the fog node with high credibility to provide services for users,this paper uses the genetic algorithm to schedule the fog computing resources,so that the three goals of time delay,communication load and service cost are optimized.The resource scheduling problem is solved.First,the multi-objective optimization problem is transformed into single-objective optimization,the preference weight is set for each target,the constraint relationship between each target is simplified,and the genetic algorithm based on linear weighting is used to obtain the resource that meets the user's demand for service preference.The scheduling scheme;secondly,the traditional non-dominated sorting genetic algorithm with the elite strategy(NSGA-II algorithm)is improved,and the dynamic adjustment model of the crossover rate and the mutation rate is established to improve the diversity of the population while maintaining the diversity of the population.The efficiency of the algorithm can finally obtain a set of optimal solution sets,which is much better than the traditional NSGA-II algorithm.The experimental results show that when the user assigns the preference weight,the optimal solution can be obtained by the genetic algorithm based on linear weighting,and the algorithm execution efficiency is high.With the increase of the single target weight,the optimization effect of the target is better,but When the preference weights tend to average,its overall optimization effect is not ideal.When the user does not specify the weight of the preference weight,an improved set of optimal solutions can be obtained by the improved non-dominated sorting genetic algorithm with elite strategy,and compared with the traditional NSGA-II algorithm and RAS-IN algorithm,except The optimization effect of the target as a whole is better,and the algorithm itself has higher efficiency.