Design And Implementation Of The Internet Of Things Plat-form Based On Cloud Native

With the continuous progress of Io T communication technologies such as 5G and narrowband Io T,various terminal devices are able to access the Io T more easily and also enhance the speed of data transmission by these devices.Several emerging fields,such as autonomous driving monitoring,cold chain logistics monitoring,etc.,are currently using Io T technologies.The Io T collection devices in these fields present the following challenges to the current Io T platform: a wide range of network communication protocols and data types,a huge number of devices,and low latency requirements for manipulation.The existing Io T platforms generally use a monolithic architecture design with high coupling,which has difficulties in expansion and compatibility and cannot cope with the above challenges.Therefore,there is an urgent need to design an open,highly scalable,and high-performing Io T platform for unified management of Io T collection device status and data.The main contents and work of this thesis are as follows.1.A device management data model is designed for the increasing variety of devices and the increasingly complex device management logic of the Io T platform.It consists of metadata,acquisition data,configuration information and other data,which can make the platform consistent in the management of different kinds of devices.When accessing a new type of device,it is only necessary to develop a targeted device access layer service for protocol conversion according to its protocol to access the platform,which simplifies the device management logic and has high maintainability and scalability.2.For the problem that the number of accessed devices and data volume are huge and cannot be handled by a single service,the overall architecture and service design specifications of the old Io T platform are optimized by referring to the design concepts and technologies in cloud native.In the overall architecture,the original monolithic architecture is split into multiple microservices using microservice design ideas,each microservice is deployed independently using container technology,communication management using service grid technology,and governance of a large number of containers using Kubernetes technology.New service design specifications were designed to require that each service in the platform be stateless or communicate with the help of message queues,and that each service not be tied to a specific device or user,thus enabling easy horizontal scaling.After testing,these designs enable the platform to provide highly concurrent,low-latency and stable services with a 100% to 200% improvement in base performance compared to Io T platforms using a monolithic architecture,and the performance can be further enhanced by adding working nodes.It can meet the challenges of accessing massive hardware devices.The cloud-native Io T platform designed and implemented in this thesis is able to provide stable,reliable and high-performance access and data processing services for a large number of devices compared to existing Io T platforms with monolithic architecture,which meets the network access requirements of different types of hardware devices and has certain research and application value,and has been used and promoted in various scenarios such as cold chain freight and port storage.