| Cloud manufacturing,edge computing and other technologies are in the stage of rapid development,and the traditional manufacturing industry is about to usher in the transformation of intelligence and digitization.However,more and more manufacturing equipment will produce geometric exponential data traffic,which makes it difficult for the traditional network infrastructure to meet the high requirements of industrial business for real-time.In the peak period of traffic,it will also cause link congestion and even reduce manufacturing efficiency and manufacturing quality,which will seriously affect the quality of service(QoS)of users.Therefore,in order to solve the above problems,in today’s manufacturing transformation,there is an urgent need for a new network architecture.Software defined network(SDN)provides a new idea to solve the problem that traditional networks are difficult to meet the QoS requirements of industrial services.SDN is a new network architecture,which separates the control function of the switch from the forwarding function,so that the switch only has the forwarding function,and the control function is only handled by the software.Developers can easily expand functions on the controller,and customize the transmission mode configuration of different business flows.Then the controller will distribute the configuration set by the developer to the switch in the form of flow table,and match the incoming data flow with the flow table,so as to have the function of specific business processing.On the new network architecture based on SDN,combined with the industrial field,a cloud-side data interactive QoS control system is proposed in this paper,which starts from three aspects: service classification,service queue management and routing calculation,so as to optimize the QoS requirements of industrial services.In terms of service classification,based on Diff Serv Model,different QoS policies are added in this paper to match the incoming data flow,and mainly distinguishes cloud services with different importance and real-time requirements;In terms of service queue management,the hierarchical token bucket(HTB)algorithm in Linux kernel is used to create a queue in each switch of the global network topology.When the data flow enters different queues after passing through the differentiated service module,each service with different real-time requirements can get reasonable bandwidth allocation;In terms of routing calculation,considering that the classical Dijkstra algorithm will make the data flow accumulate to the same shortest path,the link length and the link bandwidth are considered comprehensively in this paper,so that the improved Dijkstra algorithm can return multiple shorter paths at the same time with monitoring the change of link bandwidth in real time.When a shorter path is congested,the improved algorithm can distribute the incoming data stream to another link with less load.Finally,through building the simulation environment for SDN,the cloud-side data interactive QoS control system is tested and analyzed in this paper.The simulation results show that the system designed in this paper can distinguish different industrial services and make high priority services forward first.Besides,the improved algorithm can also improve the data transmission speed and make the network more stable. |
