GPU VMs on Azure Stack Hub

This article describes how to manage GPU VMs on Azure Stack Hub.

Partitioned GPU VM size

The NVv4-series virtual machines are powered by AMD Radeon Instinct MI25 GPUs. With the NVv4-series, Azure Stack Hub introduces virtual machines with partial GPUs. This size can be used for GPU accelerated graphics applications and virtual desktops. NVv4 virtual machines currently support only Windows guest operating system.

Size vCPU Memory: GiB Temp storage (SSD) GiB GPU GPU memory: GiB Max data disks Max NICs
Standard_NV4as_v4 4 14 88 1/8 2 4 2

Patch and Update, FRU behavior of VMs

GPU VMs undergo downtime during operations such as patch and update (PnU), and hardware replacement (FRU) of Azure Stack Hub. The following table describes the state of the VM as observed during these activities and also the manual action that the user can do to make these VMs available again post these operations.

Operation PnU - Express Update PnU - Full Update, OEM update FRU
VM state Unavailable during and post update without manual start operation Unavailable during update. Available post update with manual operation Unavailable during update. Available post update with manual operation
Manual operation If the VM needs to be made available during the update, if there are available GPU partitions, the VM can be restarted from the portal by clicking the "Restart" button. The VM needs to be restarted post update from the portal using the "Restart" button VM can't be made available during the update. After the update completes, the VM must be stop-deallocated using the "Stop" button and started backup using the "Start" button VM can't be made available during the update. After the update completes, the VM needs to be stop-deallocated using the "Stop" button and started backup using the "Start" button.

Guest driver installation

This document goes over setting up the AMD guest driver inside the NVv4 GPU-P enabled VM along with steps on how to verify driver installation.

Next steps

Azure Stack VM features