Your Browser is Out of Date

Nytro.ai uses technology that works best in other browsers.
For a full experience use one of the browsers below
E2E Deployment and Operations Guide with Switchless Networking: Microsoft HCI Solutions from Dell Technologies
Deployment Introduction Solution Overview Solution Deployment Switchless Architecture with Network ATC References GPU Integration
GPU Integration
Installing and configuring GPUs on AX platforms
AX-650 GPU configuration AX-750 GPU configuration AX-7525 GPU configuration GPU physical installation
GPU Discrete Device Assignment (DDA) on Azure Stack HCI
GPU partitioning
Host and guest driver download and install instructions
Day 1 operations Managing and monitoring clusters with Windows Admin Center Dell OpenManage Integration with Windows Admin Center Updates and maintenance References

GPU Discrete Device Assignment (DDA) on Azure Stack HCI

GPU Discrete Device Assignment (DDA) on Azure Stack HCI

  • On Azure Stack HCI servers with version 21H2 and later, there is the capability for GPU enabled clustered VMs by using DDA on supported GPUs. For the list of supported GPUs, see the latest version of the Support Matrix.

    If you are planning on using DDA on Azure Stack HCI 21H2, then an INF file needs to be installed on every host on the cluster. Instructions regarding how to install the INF file and where to download can be found at NVIDIA GPU Passthrough Support :: NVIDIA Tesla Documentation.
    Note: The INF file is not required on versions 22H2 or later.

    Before starting with DDA, we recommend installing the latest Datacenter/Tesla drivers on all your cluster nodes to avoid errors on device manager. The driver can be found at NVIDIA Download Drivers. For the Operating System field, select Windows Server 2022 even if you are using Azure Stack HCI.


    For doing DDA, you have two paths on how to implement it, Windows Admin Center (WAC) or PowerShell. Dell Technologies recommends using the Windows Admin Center to reduce user error and conform to best practices. Detailed instructions on how to perform DDA on clustered VM are provided by Microsoft and can be found here Use GPUs with clustered VMs.

    Once DDA is complete and the VM is assigned a GPU, you must install the Nvidia driver on the Guest VM for it to use the GPU. This driver can be found at NVIDIA Download Drivers. However, this time pick the correct operating system for the Guest Operating system field of the virtual machine using the DDA GPU.

    The following example shows downloading the driver for a guest VM using Windows Server 2019.


    For more instructions on installing the guest driver, see Section 4.1. Installing the NVIDIA vGPU Software Graphics Driver on Windows of NVIDIA Virtual GPU Software User Guide.

    Once the driver is installed, your VM is ready to start using the DDA GPU.