Skip to main content

Compute Setup


Running Training Jobs and Serving Models on AI Platform AKS Clusters

Managed Azure ML default​

By default, AI Platform's Traditional AI/ML journey provides an Azure Kubernetes Service (AKS) cluster as the compute target for pipelines and deployment.

AKS clusters Azure ML workspace displaying available AKS clusters.

info

Contact the AI Platform team if you have special needs for you compute target, like:

  • a virtual machine, instead of a compute cluster
  • location requirements
  • other

Example​

The development environment automatically generated by AI Platform in the Traditional AI/ML journey contains a notebook example that you can test and build upon. The notebook is set up to use the unified AKS cluster as its compute target.

notebook settings Notebook example Azure ML workspace and target details.

Extra Resources

Find Equinor specific examples in the following repositories:

Azure ML SDK materials and main repository​

Azure ML Train Model Example​

Azure ML Model Deployment​

Azure ML Pipelines Example​