---
title: Charmhub | Deploy Nova Compute Nvidia Vgpu using Charmhub - The Open Operator
  Collection
description: Deploy the latest version of Nova Compute Nvidia Vgpu on any cloud.
url: https://charmhub.io/nova-compute-nvidia-vgpu/configurations
---

# Nova Compute Nvidia Vgpu

[Aurelien Lourot](https://charmhub.io/publisher/aurelien-lourot "View all packages from Aurelien Lourot")

* [Aurelien Lourot](https://charmhub.io/publisher/aurelien-lourot "View all packages from Aurelien Lourot")

Platform:

24.04

23.10

23.04

22.10

22.04

21.10

20.04

18.04

yoga/stable 56

```
juju deploy nova-compute-nvidia-vgpu --channel yoga/stable
```

[Learn to deploy on juju >](https://juju.is/docs/juju/manage-applications)

* [force-install-nvidia-vgpu](https://charmhub.io/nova-compute-nvidia-vgpu/configurations#force-install-nvidia-vgpu)
* [vgpu-device-mappings](https://charmhub.io/nova-compute-nvidia-vgpu/configurations#vgpu-device-mappings)

[Learn about configurations >](https://juju.is/docs/juju/configuration#heading--application-configuration)

* force-install-nvidia-vgpu | boolean

  FOR TESTING ONLY. If true, the NVIDIA vGPU software will be installed and
  set up on all units regardless of the presence of NVIDIA GPU hardware. If
  false, the software will be installed and set up only on units where that
  hardware is present.
* vgpu-device-mappings | string

  YAML-formatted dict (or JSON-formatted, since JSON is valid YAML) where
  keys are vGPU types and values are physical addresses, e.g.
  .
  {'nvidia-35': ['0000:84:00.0', '0000:85:00.0'], 'nvidia-36': ['0000:86:00.0']}
  .
  See
  https://docs.openstack.org/nova/ussuri/admin/virtual-gpu.html#enable-gpu-types-compute
  and
  https://docs.openstack.org/nova/ussuri/admin/virtual-gpu.html#how-to-discover-a-gpu-type
  for more details.