Microservice Architecture

This thesis aimed to discuss the relationship between Microservice architecture and areas that are affected or affected by it, in order to provide a clear picture of how Microservice architecture works in conjunction with toolsets and pipelines in an organization. In this study two machine learning models were trained to use as microservice. The models are trained using linear regression technique and artificial neural network. Once trained APIs of both of the model is created. AN API allows an app to interact with another software. In this case the app was designed to send a request containing the inputs for both of the models and return the prediction of model using the input values. It has been concluded that a microservice architectural style is a way of building a single application as a collection of distinct services, each of which operates in its own process and interacts in lightweight ways, most often an HTTP resource API. These services are tailored to the needs of the business and can be delivered completely autonomously utilizing fully automated deployment equipment. These services are based on business capabilities and can be delivered entirely by themselves using fully automated deployment equipment. These services, which can be written in a variety of programming languages and use a variety of data storage systems, require very little centralized supervision. Today’s markets are looking for an architecture style which can easily handle scalability, development, flexibility, and easy to update and editing. Microservices architecture has been proven as one that has the required characteristics. In this architecture different services are used in an application. Usually each service in an app followed by Microservices architecture are independent from each other. This makes the app scalability easy and also easy to maintain. The services can be edited and updated individually without effecting other modules. The app created in this study proves this as both of the models can be edited individually which would not affect the other model. Same applies for scalability, more services in the app can be added without effecting the machine learning models. The main focus of this study was to analyze the characteristics of MSA by implementing an application based on it. The service-oriented architecture paradigm we saw a generation ago has been reinvented in many respects by microservices and the API-driven methodology. There are however small yet sig