HomeComputeGPU InstancesReference Content
Understanding NVIDIA GPU Cloud (NGC)
Update content

Understanding NVIDIA GPU Cloud (NGC)

Reviewed on 31 August 2023Published on 31 August 2022

NVIDIA provides access to NVIDIA GPU Cloud (NGC) through providers including Scaleway. NGC is a cloud-based platform that offers a comprehensive set of GPU-optimized software and tools for various deep learning, machine learning, and AI (Artificial Intelligence) tasks. It is designed to simplify deploying, managing, and utilizing AI and data science frameworks on NVIDIA GPUs.

Unleash the power of your applications’ deployment with Scaleway and NGC, where seamless deployment is as effortless as following these three simple steps:

  1. Deploy your Scaleway GPU Instance from the Scaleway console, the CLI tools, or the Scaleway API.1 1 You need GPU Instance quotas to be able to deploy GPU Instances.
  2. Configure your Instance by pulling the required packages from NVIDIA NGC.
  3. Run your application.

Your AI framework is ready to go.

NGC provides a repository of pre-configured containers, models, and software stacks optimized for NVIDIA GPUs. These containers contain popular AI frameworks, libraries, and software applications, allowing data scientists, researchers, and developers to quickly access and deploy these resources without requiring extensive configuration or setup. In detail, NGC provides the following contents in the NGC catalog:

  • Containers: NGC provides a range of containerized environments that encapsulate AI and deep learning frameworks such as TensorFlow, PyTorch, MXNet, etc. These containers are optimized for NVIDIA GPUs and can easily be deployed on cloud platforms or on-premises hardware.
  • Pre-trained models: NGC offers a collection of pre-trained models that cover a wide range of AI tasks, including image and speech recognition, natural language processing, and more. These models can be used as a starting point for building custom solutions.
  • Software Development Kits (SDKs): NGC provides SDKs and libraries that enable developers to integrate GPU-accelerated AI capabilities into their applications and workflows.
  • Frameworks and libraries: NGC includes popular AI frameworks like TensorFlow, PyTorch, and others, as well as specialized libraries for tasks like computer vision and deep learning.
  • Data Science tools: NGC provides tools for data preprocessing, exploration, and analysis. This allows data scientists to work more efficiently with their data.

NVIDIA closely collaborates with software developers to optimize leading AI and machine learning frameworks for peak performance on NVIDIA GPUs. This optimization significantly expedites both training and inference tasks. Software hosted on NGC undergoes scans against an aggregated set of common vulnerabilities and exposures (CVEs), crypto, and private keys.

For more information on NGC, refer to the official NVIDIA NGC documentation.