TailorNode Documentation – Using a VM with GPU

Docs: Installing NVIDIA GPU driver

Mon, 01 Jan 0001 00:00:00 +0000

This is the procedure for installing the GPU driver in the following environment:

GPU: NVIDIA A100
OS: Ubuntu20.04
GPU Driver: 550.x（LTS）

1. Installing NVIDIA GPU driver only

1.1. Download the package

Visit the NVIDIA Official Driver Download site:
NVIDIA Official Driver Download | NVIDIA
In the manual driver search field, enter A100 and select the following items:

Product Category: Data Center / Tesla
Product Series: A-Series
Product: NVIDIA A100
OS: Linux 64-bit Ubuntu 20.04

Download the required version of the driver.
Agree to NVIDIA’s terms and conditions to complete the download.

1.2. Installation

Copy the downloaded .deb file to the target server.
Execute the following commands on the server:

sudo dpkg -i <downloaded file>.deb
sudo cp /var/nvidia-driver-local-repo-ubuntu2004-550.xx.xx/nvidia-driver-local-xxxxxx-keyring.gpg /usr/share/keyrings/
## Copy and execute the part of the cp command displayed when executing the dpkg command
sudo apt update
sudo apt install nvidia-driver-550
sudo reboot

After rebooting the server, run the nvidia-smi command. If a list of GPUs is obtained, the installation is complete.
To prevent risks such as unintended suspensions that may cause the machine or GPU to become inaccessible after installation, also execute the following commands:

sudo systemctl enable multi-user.target --force
sudo systemctl set-default multi-user.target
sudo systemctl mask sleep.target suspend.target hibernate.target hybrid-sleep.target
sudo reboot

2. Install CUDA Toolkit

If CUDA is required, select the necessary version from the following site and follow the instructions to install:

CUDA Toolkit Archive