dataiku azure deployment
Hello,
I'm looking for some information, reference architecture to deploy dataiku on azure app service or container instances. we would like to set up a dev environment for testing the app.
also, any guidance , pros cons etc would be appreciated
Answers
-
Shashank Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 28 Dataiker
Below are the options for installing Dataiku on Azure:
At this point, the recommendation is to set up Dataiku on an Azure VM machine
-
JordanB Dataiker, Dataiku DSS Core Designer, Dataiku DSS Adv Designer, Registered Posts: 297 Dataiker
Hi @ashahbazi
,Great! I recommend reading through our dedicated Azure Documentation as well as our documentation on using DSS with containerized execution on AKS as a fully managed Kubernetes solution. To get a more broad understanding of how interaction works between DSS and containers, please see Elastic AI: Concepts.
Please let us know if you have any questions.
Thanks!
Jordan
-
Turribeach Dataiku DSS Core Designer, Neuron, Dataiku DSS Adv Designer, Registered, Neuron 2023 Posts: 2,161 Neuron
Here is a video explaining the Dataiku architecture:
https://www.youtube.com/watch?v=lpM4vEnDpko
More info on this page: https://knowledge.dataiku.com/latest/courses/mlops/prod-concepts/dataiku-architecture-summ.html
In a nutshell you have 3 node types: Designer, Automation and API. Neither of these nodes run on Azure App Service or Azure Container Instances, they need dedicated VMs. Having said that it is possible to execute certain Dataiku workloads in Azure Kubernetes Service (AKS).
See the following documentation to understand how that works:
https://doc.dataiku.com/dss/latest/user-isolation/reference-architectures/kubernetes.html
https://doc.dataiku.com/dss/latest/containers/managed-k8s-clusters.html
https://doc.dataiku.com/dss/latest/apinode/kubernetes/index.html