Research And Design Of Cloud Control Software For Generator Sets Based On Microservices

With the development of industrial informatization,the state vigorously promotes the policy of "promoting enterprises to accelerate the networking of industrial equipment to the cloud and the migration of business systems to the cloud",remote cloud monitoring software has been widely used in the generator sets industry.However,as the volume of equipment grows and the demand for remote monitoring gradually diversifies,the traditional monolithic software architecture has the disadvantages of high coupling,poor scalability and difficult maintenance,all of which restrict the development of remote cloud monitoring software.This topic addresses the above existing problems,researches and designs a cloud control software for generator sets based on microservice architecture,and improves the microservice deployment strategy and microservice invocation strategy of the microservice architecture during the design and implementation of the software.The main work of this thesis is as follows.1.To address the problem that most microservice deployment strategies only consider CPU and memory usage,this topic proposes an interaction-aware microservice deployment strategy based on interaction-awareness by analyzing the characteristics of interactions between microservices.The strategy first generates the interaction graph between microservice components,then calculates the interaction coefficients,and finally invokes the improved deployment strategy to deploy the microservice components.According to the test,the strategy reduces 4.3% in average response time and improves4.6% in average throughput than the default Kubernetes deployment strategy.2.To address the problem of delayed response to service requests under multi-room remote deployment of microservice invocation strategy,this topic proposes a microservice invocation strategy based on dynamic Qo S adaption by analyzing the needs and Qo S requirements of microservice chains.The strategy first establishes the workflow model,then calculates the invocation urgency and updates it in real time,and finally constructs the service path.According to the test,the strategy reduces the average response time by 3.8% and improves the number of requests completed on time by 4.7%compared to the comparison strategy proposed in the literature[23].3.The cloud control software is designed using a microservice architecture,and the cloud control software of the generator sets includes web front-end,service layer and resource layer respectively.Web front-end uses Vue.js framework to beautify front-end interface,the service layer uses Spring Cloud framework to build service back-end,and My SQL and Redis are used to store data in resource layer.The software is designed according to the requirements of four functional modules: device list,organization management,collaborative file and system management.4.Build a software testing environment for microservice architecture.According to the software quality testing standard GB/T25000.51-2016,the corresponding functional and performance tests were carried out on the software to evaluate the functional integrity and reliability of the software.The test results show that the software designed in this project meets the user’s demand for remote cloud monitoring of generator sets and conforms to the national software testing standards.