How to Install NVIDIA driver on AWS Ubuntu LTS 24.04

Installing the NVIDIA driver on an AWS EC2 instance running Ubuntu 24.04 can sometimes be challenging due to AWS’s custom environment and kernel. Although the ubuntu-drivers tool is the recommended way to install drivers on Ubuntu, we strongly recommend using apt to install NVIDIA drivers on AWS EC2 instances to ensure compatibility and a more reliable setup.

Step 1: Use apt to Install the NVIDIA Driver Directly

On AWS, using apt to install the NVIDIA driver is more effective than the default ubuntu-drivers approach, which may only install partial NVIDIA components. Installing the full driver package directly helps ensure all necessary components are in place.

To install the NVIDIA driver using apt, run:

sudo apt update
sudo apt install -y nvidia-driver-535-server

The -y flag automatically accepts any prompts during installation. The nvidia-driver-535-server package installs the complete server driver, including essential components like the Dynamic Kernel Module Support (DKMS), which ensures compatibility with AWS’s custom kernel.

Why Not Use ubuntu-drivers?

Using sudo ubuntu-drivers install --gpgpu might seem convenient, but on AWS, this method often only installs partial packages like nvidia-headless without the full driver, which can prevent the GPU from functioning correctly. To avoid these issues, we recommend directly installing the full driver package with apt, as shown above.

Step 2: Check Installed NVIDIA Packages

After installation, verify the installed NVIDIA packages with:

dpkg -l | grep nvidia

You should see a list of installed packages similar to the following:

ubuntu@ip-10-138-2-148:~$ dpkg -l | grep nvidia
ii  libnvidia-cfg1-535-server:amd64       535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA binary OpenGL/GLX configuration library
ii  libnvidia-common-535-server           535.216.01-0ubuntu0.24.04.1             all          Shared files used by the NVIDIA libraries
ii  libnvidia-compute-535-server:amd64    535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA libcompute package
ii  libnvidia-decode-535-server:amd64     535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA Video Decoding runtime libraries
ii  libnvidia-egl-wayland1:amd64          1:1.1.13-1build1                        amd64        Wayland EGL External Platform library -- shared library
ii  libnvidia-encode-535-server:amd64     535.216.01-0ubuntu0.24.04.1             amd64        NVENC Video Encoding runtime library
ii  libnvidia-extra-535-server:amd64      535.216.01-0ubuntu0.24.04.1             amd64        Extra libraries for the NVIDIA Server Driver
ii  libnvidia-fbc1-535-server:amd64       535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA OpenGL-based Framebuffer Capture runtime library
ii  libnvidia-gl-535-server:amd64         535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA OpenGL/GLX/EGL/GLES GLVND libraries and Vulkan ICD
ii  nvidia-compute-utils-535-server       535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA compute utilities
ii  nvidia-dkms-535-server                535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA DKMS package
ii  nvidia-driver-535-server              535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA Server Driver metapackage
ii  nvidia-firmware-535-server-535.216.01 535.216.01-0ubuntu0.24.04.1             amd64        Firmware files used by the kernel module
ii  nvidia-kernel-common-535-server       535.216.01-0ubuntu0.24.04.1             amd64        Shared files used with the kernel module
ii  nvidia-kernel-source-535-server       535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA kernel source package
ii  nvidia-utils-535-server               535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA Server Driver support binaries
ii  xserver-xorg-video-nvidia-535-server  535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA binary Xorg driver

These packages provide all necessary GPU functionalities, including compute support, kernel module management (DKMS), and utilities (nvidia-utils-535-server) for interacting with the GPU, including nvidia-smi.

Step 3: Reboot and Verify the Installation

After installing the driver, reboot your instance to apply changes:

sudo reboot

Once your instance is back online, check the NVIDIA driver status by running:

lsmod | grep nvidia
nvidia-smi

If the driver is correctly installed and functioning, nvidia-smi will display details about your NVIDIA GPU, including memory usage, processes, and the driver version.

Troubleshooting: Common Issues

If you encounter issues, try these troubleshooting steps:

• Driver/Library Version Mismatch: If nvidia-smi shows a version mismatch, reinstall the nvidia-driver-535-server package to ensure consistency.
• NVIDIA-SMI has failed because it couldn’t communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running: Check whether necessary packages are installed and make sure kernel headers and other packages for the AWS kernel are installed:

ubuntu@ip-10-138-2-97:~$ dpkg -l | grep nvidia
ii  libnvidia-cfg1-535-server:amd64       535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA binary OpenGL/GLX configuration library
ii  libnvidia-compute-535-server:amd64    535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA libcompute package
ii  nvidia-compute-utils-535-server       535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA compute utilities
ii  nvidia-firmware-535-server-535.216.01 535.216.01-0ubuntu0.24.04.1             amd64        Firmware files used by the kernel module
ii  nvidia-headless-no-dkms-535-server    535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA headless metapackage - no DKMS
ii  nvidia-kernel-common-535-server       535.216.01-0ubuntu0.24.04.1             amd64        Shared files used with the kernel module
ii  nvidia-kernel-source-535-server       535.216.01-0ubuntu0.24.04.1             amd64        NVIDIA kernel source package

sudo apt install nvidia-kernel-common-535
sudo apt install nvidia-dkms-535
sudo update-initramfs -u  
sudo apt install linux-headers-$(uname -r)

• Secure Boot: Secure Boot may block NVIDIA drivers. Disable Secure Boot in the BIOS if necessary.

Summary

For AWS EC2 instances, installing the NVIDIA driver using apt is the most reliable method. The ubuntu-drivers tool may not install all required components, particularly on AWS’s custom kernels. By following these steps, you can enable GPU capabilities on your AWS instance and make the most of NVIDIA-powered computations.

This guide should help you set up and verify the NVIDIA driver on your AWS Ubuntu 24.04 LTS instance using the recommended apt method for AWS.