NVIDIA launches AI Cloud Container Registry

NVIDIA today announced immediate availability of the NVIDIA GPU Cloud (NGC) container registry for artificial intelligence (AI) developers worldwide in conjunction with NVIDIA Tesla V100 GPUs going live on AWS.

NGC helps developers get started with deep learning development through no-cost access to a comprehensive, easy-to-use, deep learning software stack. The cloud-based service is available immediately to users of the just-announced Amazon Elastic Compute Cloud (Amazon EC2) P3 instances featuring NVIDIA Tesla V100 GPUs. NVIDIA plans to expand support to other cloud platforms soon.

After signing up for an NGC account, developers can download a containerised software stack that integrates and optimises a wide range of deep learning frameworks, NVIDIA libraries and CUDA runtime versions — which are kept up to date and run seamlessly in the cloud or on NVIDIA DGX supercomputer systems.

“The NVIDIA GPU Cloud democratises AI for a rapidly expanding global base of users,” said Jim McHugh, VP and GM of Enterprise Systems at NVIDIA. “NGC frees developers from the complexity of integration, allowing them to move quickly to create sophisticated neural networks that deliver the transformative powers of AI.”

Details: 

Start using the NGC container registry in three steps:

1. Sign up for a no-cost NGC account. 

2. Run an optimised NVIDIA image on the cloud service provider platform. 

3. Pull containers from NGC and get started.

Key benefits of the NGC container registry include:

• Instant access to the most widely used GPU-accelerated frameworks: Containerised software includes NVCaffe, Caffe2, Microsoft Cognitive Toolkit (CNTK), DIGITS, MXNet, PyTorch, TensorFlow, Theano and Torch, as well as CUDA for application development. 

• Tuned, tested and certified by NVIDIA for maximum performance

• Pre-integration: Containers allow users to begin deep learning jobs immediately, eliminating time-consuming and difficult do-it-yourself software integration. 

• Containers available on the NGC container registry benefit from continuous NVIDIA development, ensuring each deep learning framework is tuned for the fastest training possible on the latest NVIDIA GPUs. NVIDIA engineers continually optimise libraries, drivers and containers, delivering monthly updates.