一种基于Kubernetes的工业物联网新型调度

New Scheduler for Industrial Internet of Things Based on Kubernetes

随着工业物联网的迅速发展,需要更高的灵活性和可扩展性以实现对数据的实时处理,为此Google提出了Kubernetes容器技术。然而,该方案在处理物联网设置中的“干扰”和“能量最小化”方面存在不足。本文提出了一个新的Kubernetes控制器,该控制器用于边缘云节点上的容器管理,同时考虑碳足迹的排放、干扰和能量消耗;在多目标优化问题的基础上,采用整数线性规划方法对任务调度问题进行了求解。应用程序在可用节点上的调度时间更短,而来自其他应用程序的干扰最小,这反过来保证了最终用户的最佳性能。与现有方案相比,所提出的方法性能更优越。

With the rapid development of the industrial internet of things,greater flexibility and scalability are required to realize real-time data processing.Therefore,Google proposed Kubernetes container technology.However,this solution has shortcomings in dealing with"interference"and"energy minimization"in IoT settings.Therefore,a new Kubernetes controller is proposed here,which is used for container management on edge cloud nodes,while considering carbon footprint emissions,interference and energy consumption.On the basis of the multi-objective optimization problem,the integer linear programming method is used to solve the task scheduling problem.Since then,the scheduling time of the application on the available nodes is shorter,while the interference from other applications is minimal,which in turn ensures the best performance for the end user.Compared with the existing solutions,the proposed method has better performance.