XenCenter

GPU

XenCenter allows you to assign a dedicated graphics processing unit (GPU) or one or more virtual GPUs to a new VM during VM creation. This feature enables a VM to use the processing power of the GPU, providing better support for high-end 3D professional graphics applications. For example, CAD/CAM, GIS, and Medical Imaging applications.

For detailed information, see Configuring graphics.

Citrix Hypervisor supports Intel’s virtual GPU: a graphics acceleration solution that requires no additional hardware. It uses the Intel Iris Pro functionality embedded in some processors, and utilizes a standard Intel GPU driver installed within the VM. The motherboard must have a chipset which enables GPU functionality, for example, C226 for Xeon E3 v4 CPUs or C236 for Xeon v5 CPUs. For information about supported processors, refer to the Citrix Hypervisor Hardware Compatibility List.

The following table lists whether GPU, shared GPU, and multiple vGPU are supported for guests:

Note:

In Citrix Hypervisor 8.0 and earlier releases, you can only add one vGPU to a VM. From Citrix Hypervisor 8.1, you can add multiple vGPUs to a VM if your NVIDIA GPU supports this feature and the vGPUs are of the same type.

  GPU for Windows VMs GPU for HVM Linux VMs Shared GPU For Windows VMs Virtual GPU for Linux VMs Multiple vGPU For Windows VMs Multiple vGPU for Linux VMs
AMD YES   YES      
Intel YES   YES      
NVIDIA YES YES YES YES YES YES

You might need a vendor subscription or a license depending on the graphics card used.

When you click Add, the GPU type list displays available GPUs, supported virtual GPU types, resolution, and the maximum number of displays per virtual GPU. Select a GPU or a virtual GPU type from the list to add a GPU or a virtual GPU to the VM.

If you are using the virtual GPU feature, select Pass-through whole GPU to allow a VM to use the full processing power of the GPU. The GPU or virtual GPU selection can be modified later, if necessary. For more information, see Change VM Properties.

Note:

  • GPU Pass-through and Graphics Virtualization are only available for Citrix Hypervisor Premium Edition customers, or customers who access Citrix Hypervisor through their Citrix Virtual Apps and Desktops entitlement. For more information, see About Citrix Hypervisor Licensing.
  • There is no licensing restriction to use NVIDIA GPU pass-through for HVM Linux VMs.
  • When you allocate a GPU to HVM Linux VMs, the GPU type list displays all GPU types available on the host or the pool. However, only NVIDIA GPU pass-through is supported for HVM Linux VMs.

Enabling Intel GPU pass-through

Citrix Hypervisor supports the GPU pass-through feature for Windows 8 (32-/64-bit) VMs using an Intel integrated GPU device. This feature is supported on Haswell (Xeon E3-12xx v3) or newer CPUs that contain an Intel integrated GPU device and have a graphics-capable chipset. For more information on the supported hardware, refer to the Citrix Hypervisor Hardware Compatibility List.

When using an Intel GPU on Intel servers, the Citrix Hypervisor server control domain (dom0) have access to the integrated GPU device. In such cases, the GPU is not available for pass through. To use the Intel GPU pass-through feature on Intel servers, disable the connection between dom0 and the GPU before passing through the GPU to the VM.

To disable the connection:

  1. Select the Citrix Hypervisor host on the Resources pane.
  2. On the General tab, click Properties and then click GPU on the left pane.
  3. In the Integrated GPU pass-through section, click This server will not use the integrated GPU.

    This setting disables the connection between dom0 and the Intel integrated GPU device.

  4. Click OK.
  5. Reboot the Citrix Hypervisor server for the changes to take effect.

    The Intel GPU is now visible on the GPU type list during new VM creation and on the VM’s Properties tab.

Note:

The Citrix Hypervisor host’s external console output (for example, VGA, HDMI, DP) will not be available after disabling the connection between dom0 and the GPU.

GPU