Cloudera Docs

Create Azure NetApp Files Account, Capacity Pool and Volume

Azure NetApp Files is a service on Azure that provides a fully managed, native file share system that is accesible in the Azure cloud. The following procedure is only required if you are using Azure NetApp Files, which is not the recommended NFS service. For more information, see Quickstart: Set up Azure NetApp Files.

  1. In Azure Services, select Azure NetApp Files.
  2. In the New NetApp account, create an account.
  3. Select the existing Resource Group.
  4. In Location, select the region where the Cloudera AI environment is located.
  5. Provision the service. When deployment is done, click Go to Resource
  6. Under Storage Service → Capacity Pools, click Add pool.
  7. For Capacity, enter 4 TiB.
    4 TiB is the minimum size for a capacity pool in Azure NetApp Files. We recommend a Service level of Premium or higher to ensure adequate I/O performance.
  8. Click Create.
    In the next steps, create a Volume.

Create a Volume

In the Azure portal, you create an Azure NetApp Files volume to provide a file system within the Capacity Pool.

  1. In the NetApp account screen, click Volumes in the menu on the left, the click Add Volume.
  2. For Quota, set a value in GB. You can set an initial value, then increase it to add additional workbenches, without having to resize the pool itself. We recommend an initial volume capacity of at least 100GB in order to ensure adequate I/O performance.
  3. Confirm the value in the Delegated Subnet: the only option should be the ‘netapp’ subnet, because that is the one that was delegated to this service.
  4. On the Protocol tab, choose the protocol version. Version 3 is supported by Cloudera AI. Ensure that “Root Access” in the “Export policy” section is set to “On”, and “Access” is set to “Read & Write”.
  5. On the Review and Create tab, click Create. The volume is deployed after several minutes.
  6. After the volume is deployed, obtain the Mount path.
  7. Create a Linux VM in the subnet in which the Cloudera AI Workbench will be created, and verify that the NFS volume can be mounted successfully in the VM. This step ensures that there are no connectivity issues between the NFS server and hosts running in the subnet.
  8. Create a sub-directory to be used for the Cloudera AI Workbench project files. For example, if the NFS export path is 10.102.47.132:/netapp-vol and is mounted as /mnt/netapp-vol, run sudo mkdir /mnt/netapp-vol/ml-workbench-1; sudo chown 8536:8536 /mnt/netapp-vol/ml-workbench-1 to ensure the directory is owned by the cdsw user ID and group ID.
  9. When prompted for the NFS path in the Cloudera AI control plane, enter the path as 10.102.47.132:/netapp-vol/ml-workbench-1.
  10. The Linux VM created above can be deleted if there are no issues with mounting the NFS share.