DataIKU Deployment and Installation in Azure Kubernetes

[mettuharish]

Hello, 

I'm looking for some information, reference architecture to deploy dataiku on Azure Kubernetes. We would like to set up a dev environment for testing the app.

also, any guidance , pros cons etc would be appreciated .

Also looking for documentation and steps on how to deploy the app on Azure K8's.

---

Operating system used: Redhat

[Turribeach]

The first thing to understand is that Dataiku, in general terms, does not run in Kubernetes. I say that because the main node components, Designer and Automation, do not run in Kubernetes. There are however certain things that can in Kubernetes:

1. API Node: https://doc.dataiku.com/dss/latest/apinode/kubernetes/index.html
2. Certain workloads: https://doc.dataiku.com/dss/latest/user-isolation/reference-architectures/kubernetes.html
3. Code Studios: https://doc.dataiku.com/dss/latest/code-studios/index.html

My advice will be that you do not look at any of the above until you have some actual and real ML use cases that need the above components AND more importantly the use case will benefit from using Kubernetes.