The Research And Implementation Of Edge-based Microservice Architecture For Industrial Iot

Industrial Internet of Things has become an important support of industrial digitization,networking and intelligence.However,with the rapid development of Industrial applications,the collection,aggregation and computation of large-scale sensing data make the network bandwidth and service response face challenges,which also become the bottleneck for the development of Industrial Internet of Things.For that reason,this thesis raises a microservice architecture based on edge computing in industrial environment,then implements the architecture and optimize the performance in a real industrial scene.Different from the traditional architecture of asynchronous data transmission and processing in industrial Internet,this architecture proposed by this thesis used computation offloading to extend the operation to the edge and data source devices,which makes data transmission and computing tightly coupled.The thesis mainly includes the following aspects of research:1.Designed edge-based microservice process,network structure and interaction logic for the industrial scenario,and the architecture of edge microservice is constructed to make data transmission and computing more flexible.2.The control method of microservice life cycle is proposed.Furthermore,in order to manage industrial service data,this architecture designs and implements data feedback strategy to keep data conversion and execution visible.After classifying the industrial service based on microservice network,the thesis builds a standardized microservice set.3.The thesis implements the whole system based on Spring MVC framework and Docker,and deployed in industrial internet in Jiangsu Nangang.The results shows that the overall performance of the architecture has an obvious improvement compared with traditional computing paradigms,especially in bandwidth and response time.