Available on Amazon EC2 P3 instances, the containerized software stack offers access to deep learning frameworks, libraries, and more.
The NVIDIA GPU Cloud (NGC) container registry, announced via a press release on Thursday, could make it easier for developers to get started working with artificial intelligence (AI) and deep learning technologies.
The NGC was originally announced back in May 2017, but it is now generally available. The registry will be available on Amazon EC2's P3 instances, leveraging the NVIDIA Tesla V100 GPUs, the release said. The NVIDIA Tesla V100 GPUs are optimized for AI, HPC, and graphics workloads.
To access the features of the registry, users with an account (sign up here) can simply download a containerized software stack that provides access to specific NVIDIA libraries, deep learning frameworks, and CUDA runtime versions, the release said.
SEE: IT leader's guide to the future of artificial intelligence (Tech Pro Research)
"Easy-to-use containers allow users to begin deep learning jobs immediately, eliminating time-consuming and difficult do-it-yourself software integration," the release said. The containers will also receive consistent updates and optimizations from NVIDIA.
Developers using NGC will have access to frameworks like NVCaffe, Caffe2, Microsoft Cognitive Toolkit (CNTK), DIGITS, MXNet, PyTorch, TensorFlow, Theano, and Torch, the release said.
"The NVIDIA GPU Cloud democratizes AI for a rapidly expanding global base of users," Jim McHugh, vice president and general manager of Enterprise Systems at NVIDIA, said in the release. "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."
Also on Thursday, NVIDIA announced a partnership with Taiwan's Ministry of Science and Technology (MOST) to boost AI development in the commercial sector, and support Taiwan's AI Grand Plan to create AI-related businesses in the country, a separate press release said.
The announcement came during NVIDIA's GPU Technology Conference in Taiwan, where it was revealed that a supercomputer based on NVIDIA technology would be built by the National Center for High-Performance Computing, the release said.
"AI is the key to igniting Taiwan's next industrial revolution, building on the long-established strength of our IT manufacturing capabilities," Liang-Gee Chen, minister of science and technology for Taiwan, said in the release. "Our focus is on drawing academics, industry and young talent into our AI Grand Plan to create an ecosystem based on AI innovation."
The 3 big takeaways for TechRepublic readers
- The NVIDIA GPU Cloud (NGC) container registry offers developers a containerized software stack that includes a host of technologies needed to run deep learning and AI workloads.
- The NGC uses NVIDIA Tesla V100 GPUs running on Amazon EC2's new P3 instances, optimized for AI.
- NVIDIA is also partnering with Taiwan to help advance AI efforts in its commercial sector, while the National Center for High-Performance Computing aims to build a supercomputer with NVIDIA technology.
- Special report: How to implement AI and machine learning (free PDF) (TechRepublic)
- AWS says its new monster GPU array is 'most powerful' in the cloud (ZDNet)
- The Complete Machine Learning Bundle (TechRepublic Academy)
- How Microsoft plans to turn Azure into an 'AI cloud' (ZDNet)
- New Microsoft cloud service uses AI to find bugs in your code (TechRepublic)