Posted on
• Administration

Installing CUDA across RHEL, Ubuntu, and openSUSE

Author
  • User
    Linux Bash
    Posts by this author
    Posts by this author

Installing CUDA on Linux: A Step-by-Step Guide for RHEL, Ubuntu, and openSUSE

The CUDA (Compute Unified Device Architecture) platform from NVIDIA is a powerful toolset that enables dramatic increases in computing performance by harnessing the power of the graphics processing unit (GPU). Whether you're involved in data science, machine learning, or other intensive computational tasks, installing CUDA can substantially accelerate your processes. This guide provides detailed instructions on how to install CUDA on three popular Linux distributions: Red Hat Enterprise Linux (RHEL), Ubuntu, and openSUSE.

Common Prerequisites

Before diving into the specific steps for each distribution, ensure your system meets the following: 1. An NVIDIA GPU with a CUDA compute capability of 3.5 or higher. 2. Ensure that your system is updated and that you have the necessary rights (typically root access) to install software.

Ubuntu (Using apt)

Ubuntu, being one of the most popular Linux distributions, provides a straightforward method for installing CUDA through its package manager, apt.

  1. Add NVIDIA package repositories First, you'll need to add the NVIDIA package repositories that include the CUDA toolkit.

    wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/cuda-ubuntu1804.pin
sudo mv cuda-ubuntu1804.pin /etc/apt/preferences.d/cuda-repository-pin-600
sudo apt-key adv --fetch-keys http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub
sudo add-apt-repository "deb http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/ /"
sudo apt-get update

  2. Install CUDA

    sudo apt-get -y install cuda

  3. Configure your environment Add the following lines to the end of your .bashrc file:

    export PATH=/usr/local/cuda-11.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-11.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

  4. Reboot Reboot your system to apply all changes:

    sudo reboot

Red Hat Enterprise Linux (RHEL) (Using dnf)

For RHEL, the recommended way is to use the dnf package manager. The process differs slightly from that used in Ubuntu.

  1. Enable the CUDA repository

    sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo

  2. Install CUDA

    sudo dnf clean all
sudo dnf -y module install nvidia-driver:latest-dkms
sudo dnf -y install cuda

  3. Configure your environment as described for Ubuntu

  4. Reboot your system

openSUSE (Using zypper)

In openSUSE, zypper is used as the package manager to handle installations.

  1. Add the NVIDIA CUDA repository

    sudo zypper addrepo https://developer.download.nvidia.com/compute/cuda/repos/opensuse15/x86_64/cuda-opensuse15.repo

  2. Refresh and install CUDA

    sudo zypper refresh
sudo zypper install cuda

  3. Configure your environment as per the instructions given for Ubuntu

  4. Reboot your system

Post-Installation

After installation, it's a good practice to verify that the installation was successful. Run:

nvidia-smi

This command should display information about the NVIDIA driver and GPU device.

Conclusion

Installing CUDA across different Linux distributions involves similar steps: adding the NVIDIA repository, installing CUDA, configuring the environment, and rebooting the system. By following these steps, you can equip your Ubuntu, RHEL, or openSUSE system with CUDA and take advantage of GPU-accelerated computing for your applications. Enjoy the power of parallel GPU computing!

Related posts