Nvidia, the global leader in AI, unveiled DGX Cloud with Oracle Cloud Infrastructure (OCI) today to make Advanced Generative AI Training at scale more accessible. DGX Cloud is a GPU-powered, multi-node, deep learning platform that is based on AI optimized hardware, software, and managed services.The new offering provides data scientists with a comprehensive and powerful cloud platform for developing and running AI workloads. DGX Cloud on OCI accelerates the deployment and runtime of deep learning frameworks for training AI models quickly and efficiently.
The platform gives users access to Nvidia’s state-of-the-art hardware components, including DGX servers, volta-based Nvidia Tesla V100 GPUs and 16GB HBM2 memory. In addition, DGX Cloud on OCI is pre-integrated with the Nvidia NV Series, the world’s first GPUs designed for AI and high-performance computing. This makes it easy for users to deploy and manage their AI and HPC workloads.
DGX Cloud on OCI provides users with access to a range of managed services, including a Container Registry, Deep Learning SDK, and Theta Compute for optimizing their GPU-accelerated workflows. Furthermore, it has been designed to support both the NVIDIA deep learning Frameworks as well as Oracle modern cloud data engineering, analytics, and AI services.
The new platform is expected to open the door to a much wider range of development and data science use cases, while offering a secure, exceedingly powerful platform for AI and HPC workloads. By deploying on Nvidia DGX Cloud on OCI, users can now quickly and easily access the most advanced hardware and software for AI training, allowing them to reduce time to market and accelerate time to insights.
With the goal of democratizing AI and HPC, Nvidia and Oracle Cloud Infrastructure have partnered to bring the power of deep learning to the world. DGX Cloud on OCI provides users with the opportunity to harness the latest in AI technology, allowing them to build and scale their models more rapidly and accurately.
