Your challenge is to deploy a python workload on Kubernetes.
The goal is to deploy the application from the app directory to a
local Kubernetes cluster.
Before running the application, please re-implement the square() function
as described in the python file.
Two versions of the python application should be available, with
different resource constraints. Both should be available from within
the default namespace under specific domain names:
- A version that outputs the square series (
SERIES_VERSION=square)- Resource limits: 100m CPU, 100Mi RAM
- Available as
http://square
- A version that outputs the fibonacci series (
SERIES_VERSION=fibonacci)- Resource limits: 200m CPU, 200Mi RAM
- Available as
http://fibonacci
To get started with the application (in this case the square version):
pip install -r app/requirements.txt
SERIES_VERSION=square uvicorn --host 0.0.0.0 --app-dir app main:app
To test the application:
curl localhost:8000/series?n=5
{"square": 25}
You may improve upon or modify the elements that are already present, as you see fit.
We give roughly equal weight to each of the following:
- Functionality
- Simplicity
- Readability
- Extensibility
- Maintainability
- Observability
Please don't stress if your solution isn't perfect, or is incomplete. With the given time constraints, you won't be able to tick all the boxes above. This challenge is intended as an exercise in tradeoffs. We want to understand how you approach platform building.