Deploying AI Models as Microservices
Dattaraj Jagdish Rao · 2019
In this chapter, the author provides some more details of building applications using Kubernetes. He begins by building a simple microservice application and then packaging it into a container. The idea of microservices is that the application is self-contained so it can be deployed and scaled independently as a container instance. First, the application will only show a simple message by reading a text string. Later some processing is performed on that text string. The author also builds a simple web application using Python, packages it as a Docker container, and deploys to a Kubernetes cluster. Just like with a regular web app, he creates a deployment for the application containing an artificial intelligence model. The author also updates this application to add code to invoke a Deep Learning Natural Language Processing model and display results on a web page.