> ## Documentation Index
> Fetch the complete documentation index at: https://anaconda.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Azure Kubernetes Service

This guide offers recommended configurations and settings unique to Azure Kubernetes Service (AKS). These should be
used to augment the generic requirements offered on our [primary requirements page](/docs/data-science/5.7.1/environment-prep/byok8s-prep).

## Instance types

* Minimum: `Standard_D8s_v4`
* Recommended: `Standard_D16s_v4`

## Storage

Unfortunately, we have found that Azure’s built-in, managed Network File System (NFS) service, [Azure Files NFS](https://docs.microsoft.com/en-us/azure/storage/files/storage-files-how-to-create-nfs-shares), does not provide an acceptable performance level for use with Data Science & AI Workbench. We have not yet had the opportunity to evaluate [Azure NetApp Files](https://azure.microsoft.com/en-us/services/netapp/).

For this reason, Anaconda recommends creating a separate Virtual Machine for hosting NFS storage. Follow the recommendations offered in [this document](/docs/data-science/5.7.1/environment-prep/env-specific-reqs/nfs), with these Azure-specific recommendations:

* The `Standard_D4s_v3` machine type is suitable for this purpose.
* Azure tightly couples disk size and IOPS performance. Anaconda recommends a minimum disk size of 1 TiB to ensure good performance.

<Note>
  This server can be the administration server as well.
</Note>

## Network

Azure offers [two different networking options](https://docs.microsoft.com/en-us/azure/aks/concepts-network) for AKS clusters. Both approaches are compatible with Workbench, so the determination depends upon your larger networking needs.

<Note>
  AKS uses a LoadBalancer which by default sets TCP idle timeouts to 4 minutes, and enables TCP resets. This may affect any user workload that depends on a continuous TCP connection.
</Note>

## GPUs

Please see [this Azure guide](https://docs.microsoft.com/en-us/azure/aks/gpu-cluster) for adding GPU resources to your AKS cluster.