NCA Containerization & NGC Catalog 4 — Questions and Answers
Question 1: Which NGC resource type provides pre-trained model weights that can be downloaded and used directly for inference or fine-tuning?
- NGC Models (Correct answer)
- NGC Blueprints
- NGC Scripts
- NGC Helm Charts
Correct answer: NGC Models
NGC Models are pre-trained model checkpoints available in the NGC Catalog that can be downloaded for direct inference or as starting points for fine-tuning.
Question 2: What is the primary benefit of using multi-stage Docker builds when creating NGC-based custom containers?
- It enables GPU pass-through without the NVIDIA Container Toolkit
- It reduces final image size by separating build dependencies from runtime artifacts (Correct answer)
- It automatically publishes the image to the NGC Private Registry
- It enables containers to run without root privileges by default
Correct answer: It reduces final image size by separating build dependencies from runtime artifacts
Multi-stage Docker builds allow build tools and intermediate files to be excluded from the final image, significantly reducing its size and attack surface.
Question 3: When using NVIDIA Multi-Instance GPU (MIG) with containers, which environment variable specifies the MIG device instance to expose?
- NVIDIA_MIG_DEVICE
- CUDA_VISIBLE_DEVICES
- NVIDIA_VISIBLE_DEVICES (Correct answer)
- MIG_INSTANCE_ID
Correct answer: NVIDIA_VISIBLE_DEVICES
NVIDIA_VISIBLE_DEVICES is used to specify which GPU or MIG device instance (e.g., MIG-GPU-<uuid>/0/0) is exposed to the container.
Question 4: What type of NGC Catalog resource are 'NVIDIA NIM' microservices classified as?
- Helm Charts
- AI Models
- Container Images (Correct answer)
- Jupyter Notebooks
Correct answer: Container Images
NVIDIA NIM (NVIDIA Inference Microservices) are packaged as optimized container images in the NGC Catalog, making AI model inference deployable via standard container workflows.
Question 5: Which NGC CLI subcommand is used to download a model artifact from the NGC Catalog to local storage?
- ngc registry model download-version (Correct answer)
- ngc model pull
- ngc catalog fetch model
- ngc download --type model
Correct answer: ngc registry model download-version
The `ngc registry model download-version <org>/<model>:<version>` command downloads a specific model version from the NGC Catalog.
Question 6: What is the significance of the 'NVIDIA Deep Learning Frameworks' containers on NGC compared to upstream framework containers?
- They include a paid license key that must be activated before use
- They are optimized with NVIDIA-specific performance patches, cuDNN tuning, and mixed-precision support not present in upstream images (Correct answer)
- They replace standard Python with NVIDIA's proprietary interpreter
- They run exclusively on NVIDIA DGX systems and cannot be used on other hardware
Correct answer: They are optimized with NVIDIA-specific performance patches, cuDNN tuning, and mixed-precision support not present in upstream images
NGC framework containers include NVIDIA-specific optimizations such as cuDNN integration, TensorRT support, AMP tuning, and performance patches that are not in community upstream images.
Question 7: In the context of NGC containers, what does 'container image digest' provide compared to a mutable image tag?
- Faster pull speeds due to compressed layers
- An immutable reference to a specific image version ensuring reproducibility (Correct answer)
- Automatic GPU driver compatibility checking
- A digital signature proving NVIDIA authored the image
Correct answer: An immutable reference to a specific image version ensuring reproducibility
An image digest (SHA256 hash) uniquely and immutably identifies a specific image manifest, ensuring you always get the exact same image regardless of whether the tag has been updated.
NCA Containerization & NGC Catalog 4
Which NGC resource type provides pre-trained model weights that can be downloaded and used directly for inference or fine-tuning?