NCP-AII Question 111
Select 4You are tasked with setting up an environment to run containerized AI workloads on a Linux server equipped with an NVIDIA GPU. As part of the setup, you need to install the NVIDIA Container Toolkit. Which of the following steps are required to successfully install the toolkit?
- A
Add the NVIDIA Container Toolkit repository to your package manager.
- B
Install the drivers for your NVIDIA GPU if they are not already installed.
- C
Run the 'nvidia-container-cli' command to initialize the toolkit.
- D
Install the NVIDIA Container Toolkit package using your system's package manager.
- E
Configure the Docker runtime to use 'nvidia' as the default runtime.
Show answer and explanation
Correct answers: A, B, D, E
Explanation
To install the NVIDIA Container Toolkit, you need to add its repository to your package manager, ensure NVIDIA GPU drivers are installed, install the toolkit package, and configure the Docker runtime to use 'nvidia'. Running 'nvidia-container-cli' is not part of the installation process but is used for low-level interactions with the toolkit after installation.
- A. Correct.
Correct: Adding the NVIDIA Container Toolkit repository ensures you can fetch the required packages from NVIDIA's official source.
- B. Correct.
Correct: The NVIDIA GPU drivers must be installed for the container toolkit to interface with the hardware properly.
- C. Incorrect.
Incorrect: While 'nvidia-container-cli' is part of the toolkit, you do not need to run this command to initialize the installation.
- D. Correct.
Correct: The container toolkit package must be installed using the package manager to set up the required binaries and libraries.
- E. Correct.
Correct: Configuring the Docker runtime to use 'nvidia' as the default runtime is necessary for Docker to leverage GPU resources when running containers.