Installing GRID® drivers for virtual workstations

Linux Windows

After you create an NVIDIA® GRID® virtual workstation, you need to install a GRID driver to support the GPUs.

For information about creating an NVIDIA® GRID® virtual workstation, see Creating VMs with attached GPUs.

Before you begin

If you want to use the command-line examples in this guide, do the following:
1. Install or update to the latest version of the gcloud command-line tool.
2. Set a default region and zone.
If you want to use the API examples in this guide, set up API access.

Install requirements

Install the build tools for your Linux version.

Debian/Ubuntu

sudo apt install build-essential

CentOS/RHEL

Run the following commands:

sudo yum group install "Development Tools"

```
elfutils-libelf-devel
```

SLES

SLES 12

sudo zypper install -t pattern devel_basis

sudo zypper install gcc8

SLES 15

sudo SUSEConnect -p sle-module-development-tools/15/x86_64

sudo zypper install gcc8

Install the Linux kernel headers. See Verify the System has the Correct Kernel Headers and Development Packages Installed in the Nvidia documentation.

Install GRID drivers

Linux

Download the GRID driver. For a full list of NVIDIA® GRID® drivers that you can use on Compute Engine, see GRID® drivers for virtual workstations.

For example, to download the NVIDIA 470.63 driver, run the following command:
```
curl -O https://storage.googleapis.com/nvidia-drivers-us-public/GRID/GRID13.0/NVIDIA-Linux-x86_64-470.63.01-grid.run
```
Start the installer.

For example, to start the NVIDIA 470.63 driver, run the following command:
```
sudo bash NVIDIA-Linux-x86_64-470.63.01-grid.run
```
During the installation, choose the following options:
- If you are prompted to install 32-bit binaries, select Yes.
- If you are prompted to modify the x.org file, select No.

Windows Server

Depending on your version of Windows Server, download the required NVIDIA® GRID® drivers. For example, to download the NVIDIA 471.68 drivers, choose one of the following:
- Windows Server 2016 and 2019
- Windows Server 2012 R2 (64-bit)
For a full list of NVIDIA® GRID® drivers that you can use on Compute Engine, see GRID® drivers for virtual workstations.
Run the installer, and choose the Express installation.
After the installation is complete, restart the VM. When you restart, you are disconnected from your session.
Reconnect to your instance using RDP or a PCoIP client.

Verify that the GRID driver is installed

Linux

Run the following command:

nvidia-smi

The output of the command looks similar to the following:

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.63.01    Driver Version: 470.63.01    CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| 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  Tesla T4            On   | 00000000:00:04.0 Off |                    0 |
| N/A   71C    P8    21W /  70W |      0MiB / 15109MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

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

Windows Server

Connect to your Windows instance using RDP or a PCoIP client.
Right-click the desktop, and select NVIDIA Control Panel.
In the NVIDIA Control Panel, from the Help menu, select System Information. The information shows the GPU that the VM is using, and the driver version.

What's next?

Learn more about GPUs on Compute Engine.
To monitor GPU performance, see Monitoring GPU performance.
To handle GPU host maintenance, see Handling GPU host maintenance events.
To optimize GPU performance, see Optimizing GPU performance.