Skip to content

Creating new servers with virtual GPU support

In select Cleura Cloud Public Cloud and Compliant Cloud regions, you can instantiate virtual machines — or simply servers — with virtual GPU (vGPU) support. For that, you use the openstack CLI utility.


Although not strictly necessary, it is certainly helpful to have some familiarity with creating servers in Cleura Cloud, especially with the OpenStack CLI.

Instantiating a vGPU server

Before you begin, please make sure to source your openstack RC file first.

As we discuss in the guide on creating servers, an openstack command for instantiating a new server looks like this:

openstack server create \
    --flavor $FLAVOR_NAME \
    --image $IMAGE_NAME \
    --boot-from-volume $VOL_SIZE \
    --network $NETWORK_NAME \
    --security-group $SEC_GROUP_NAME \
    --key-name $KEY_NAME \
    --wait \

For servers with vGPU support, the only acceptable value for IMAGE_NAME is Ubuntu 22.04 NVGRID x86_64. Additionally, the value for FLAVOR_NAME should be a flavor prefixed by g. To see all such flavors, you can type something like the following:

$ openstack flavor list -c Name -f value | grep "^g\."

Taking all of the above into account, an openstack command for creating a new vGPU server could look like the following:

openstack server create \
    --flavor "g.8c32gb50" \
    --image "Ubuntu 22.04 NVGRID x86_64" \
    --boot-from-volume 32 \
    --network cc-net \
    --security-group default \
    --key-name mykey \
    --wait \

Before you check whether the server does have virtual GPU support, we recommend assigning a floating IP to it.

Checking the status of vGPU support

From your local computer, connect to the new remote server via SSH. Type, for instance:

ssh ubuntu@<floating_ip_of_new_server>

Alternatively, connect to it using the openstack utility:

openstack server ssh my-vgpu-server -- -l ubuntu

Upon successful connection to the server and logging into the ubuntu account, in your terminal window you will see something like this:

Welcome to Ubuntu 22.04.2 LTS (GNU/Linux 5.15.0-78-generic x86_64)

 * Documentation:
 * Management:
 * Support:

  System information as of Sun Aug 13 06:11:14 UTC 2023

  System load:  0.0                Processes:             159
  Usage of /:   17.2% of 29.90GB   Users logged in:       0
  Memory usage: 1%                 IPv4 address for ens3:
  Swap usage:   0%

 * Strictly confined Kubernetes makes edge and IoT secure. Learn how MicroK8s
   just raised the bar for easy, resilient and secure K8s cluster deployment.

Expanded Security Maintenance for Applications is not enabled.

0 updates can be applied immediately.

Enable ESM Apps to receive additional future security updates.
See or run: sudo pro status

The list of available updates is more than a week old.
To check for new updates run: sudo apt update

Last login: Sun Aug 13 05:48:57 2023 from

Your first clue regarding proper vGPU support should be in the output of the lspci command:

$ lspci

00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] (rev 02)
00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II]
00:01.1 IDE interface: Intel Corporation 82371SB PIIX3 IDE [Natoma/Triton II]
00:01.2 USB controller: Intel Corporation 82371SB PIIX3 USB [Natoma/Triton II] (rev 01)
00:01.3 Bridge: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 03)
00:02.0 VGA compatible controller: Cirrus Logic GD 5446
00:03.0 Ethernet controller: Red Hat, Inc. Virtio network device
00:04.0 SCSI storage controller: Red Hat, Inc. Virtio SCSI
00:05.0 Communication controller: Red Hat, Inc. Virtio console
00:06.0 3D controller: NVIDIA Corporation GA102GL [A10] (rev a1)
00:07.0 Unclassified device [00ff]: Red Hat, Inc. Virtio memory balloon
00:08.0 Unclassified device [00ff]: Red Hat, Inc. Virtio RNG

Notice the line that says 3D controller: NVIDIA Corporation; that’s good! To get much more information regarding vGPU support, use the NVIDIA system management interface utility. That would be nvidia-smi, which is installed by default in the Ubuntu 22.04 NVGRID x86_64 image:

$ nvidia-smi

| NVIDIA-SMI 525.105.17   Driver Version: 525.105.17   CUDA Version: 12.0     |
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|   0  NVIDIA A10-12C      On   | 00000000:00:06.0 Off |                    0 |
| N/A   N/A    P8    N/A /  N/A |      0MiB / 12288MiB |      0%      Default |
|                               |                      |             Disabled |

| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|  No running processes found                                                 |

Among other pieces of information, you get the versions of the nvidia-smi utility, the NVIDIA driver, and the CUDA API.

Last update: 2023-08-18
Created: 2023-08-13
Authors: Christos Varelas, Florian Haas