Research On Service Classification And Discovery Based On Microservice Architecture

At present,the microservice architecture has become the key technology of distributed applications,and the services under this architecture have great availability and scalability,which can improve the stability of the whole system.More and more enterprises think that this architecture is a good solution for building enterprise applications.In recent years,the components and technologies of microservice framework have become more and more mature,and some enterprises have split and reorganized their applications,realizing the transition to microservice architecture.With the refinement of the split granularity,the number of microservices shows an explosive growth.However,most enterprises still use manual documents to assist retrieval,which leads to the low efficiency of microservice discovery without classification.On the other hand,there are related components of the service registry in the microservice architecture,but the service discovery mechanism provided by these components can only solve the problem of dynamic addressing and does not provide consumers with a unified and reliable retrieval method.Therefore,it is an urgent problem to provide a reliable model for microservice discovery.Based on the above analysis,this paper puts forward the concept of microservice preregistration center and constructs the corresponding model.The proposed model consists of three parts:microservice information integration module,classification module and discovery module,and microservice profile as information input,in which microservice profile is used to manage microservice related information,such as microservice name,description,etc.,these information is also the premise of microservice classification and discovery.The information integration module is used to parse the microservice configuration file,pass the information needed for classification into the classification module,integrate the classification results received into the relevant information,and finally store them in the microservice information database for microservice discovery;the classification module uses the classification model proposed in this paper to classify microservices,and the results obtained can be used for microservice discovery to improve the discovery efficiency and accuracy.The discovery module realizes multi-scene retrieval of microservices through string matching and semantic similarity matching,which provides consumers with a reliable way of microservice discovery.The experimental results show that the pre-registry model proposed in this paper can be used for efficient microservice discovery,solves the current problem of inefficient discovery,and has good practical value.In addition,the classification model proposed in this paper has high accuracy and plays a key role in the efficient discovery of microservices.