The Ubuntu 22.04 and 24.04 64-bit images with pre-installed NVIDIA GPU drivers are high-performance public images optimized for AI development and deep learning. These images are pre-installed with NVIDIA GPU drivers, CUDA, the Docker engine, and the NVIDIA Container Toolkit. They are available out-of-the-box and allow you to quickly deploy a containerized GPU environment for large model training and inference tasks. This simplifies the configuration of underlying dependencies and improves the efficiency of AI application development and deployment.
Pre-configured software information
The pre-installed drivers and software in these public images are detailed below:
Kernel version, driver, and software information | Ubuntu 22.04 64-bit image with pre-installed NVIDIA GPU driver | Ubuntu 24.04 64-bit image with pre-installed NVIDIA GPU driver |
Operating system kernel version | 5.15.0-161-generic | 6.8.0-88-generic |
NVIDIA GPU driver version | 570.195.03 | 570.195.03 |
NVIDIA Container Toolkit | 1.17.8 | 1.17.8 |
CUDA version | 12.8 | 12.8 |
cuDNN version | 9.8.0.87 | 9.8.0.87 |
NCCL | v2.27.7-1 | v2.27.7-1 |
OpenMPI | 4.1.3 | 4.1.3 |
Python 3 | 3.10.12 | 3.12.3 |
Docker | 29.1.3 | 29.1.3 |
Keentune performance tuning software Disabled by default. | Support | Support |
OFED and elastic Remote Direct Memory Access (eRDMA) | Support | Supported |
For information about the drivers and software in previous image versions, see Ubuntu Public Image Release Notes.
Ubuntu 22.04 64-bit image with pre-installed NVIDIA GPU driver
Supported instance families
gn7e, gn7s, gn7i, gn6v, gn6i, gn6e, gn5, and gn5i
ebmgn7e, ebmgn7i, ebmgn6v, ebmgn6i, and ebmgn6e
ebmgn7ix, ebmgn7ex
gn8is, ebmgn8is, gn8v, and ebmgn8v
System environment variable configurations
/etc/profile.d/nccl.sh
export NCCL_HOME=/usr/local/nccl export LD_LIBRARY_PATH=${NCCL_HOME}/lib:$LD_LIBRARY_PATH/etc/profile.d/openmpi.sh
export MPI_HOME=/usr/local/openmpi export LD_LIBRARY_PATH=${MPI_HOME}/lib:$LD_LIBRARY_PATH export PATH=${MPI_HOME}/bin:$PATH/etc/profile.d/cuda.sh
export PATH=/usr/local/cuda/bin:$PATH export CUDA_HOME=/usr/local/cuda export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
Ubuntu 24.04 64-bit image with pre-installed NVIDIA GPU driver
Supported instance families
gn7e, gn7s, gn7i, gn6v, gn6i, gn6e, gn5, and gn5i
ebmgn7e, ebmgn7i, ebmgn6v, ebmgn6i, and ebmgn6e
ebmgn7ix and ebmgn7ex
gn8is, ebmgn8is, gn8v, and ebmgn8v
System environment variable configurations
/etc/profile.d/nccl.sh
export NCCL_HOME=/usr/local/nccl export LD_LIBRARY_PATH=${NCCL_HOME}/lib:$LD_LIBRARY_PATH/etc/profile.d/openmpi.sh
export MPI_HOME=/usr/local/openmpi export LD_LIBRARY_PATH=${MPI_HOME}/lib:$LD_LIBRARY_PATH export PATH=${MPI_HOME}/bin:$PATH/etc/profile.d/cuda.sh
export PATH=/usr/local/cuda/bin:$PATH export CUDA_HOME=/usr/local/cuda export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
FAQ
How do I enable the Keentune tuning tool for an Ubuntu image with a pre-installed NVIDIA GPU driver?
You can enable the tool by following these steps. The changes take effect after you restart the operating system.
systemctl stop tuned
systemctl disable tuned
systemctl start keentune-target
systemctl enable keentune-target
systemctl enable keentuned
systemctl start keentuned
keentune profile set cpu_ubuntu_common.profileTo disable Keentune, run the keentune profile rollback command. The change takes effect after you restart the operating system.
What should I note when using an Ubuntu image with a pre-installed NVIDIA GPU driver in an ACK cluster?
For more information, see the Container Service for Kubernetes documents Create a custom image from an ECS instance and use the image to create a node and Usage notes and high-risk operations.