Github Repo
An exhaustive explanation of this system is within the repo’s README.md itself, linked below:
https://github.com/Don-Irwin/fastapi_ml_k8_k6_gh_actions
Background
This is a Rest API serving a Prediction ML model from Docker containers within a Kubernetes (Minikube) Cluster.
A Redis Cache is used to lessen the load on the ML API.
System load presented through a Graphana dashboard.
A simple web application is exposed to consume the FastAPI rest service serving the ML model.
Github actions are used to test the quality of the deploy any time the dev branch is merged into the main branch.