The OCI Supercluster, which is connected to NVIDIA networking, powers the NVIDIA AI Foundations and DGX Cloud.
Oracle and NVIDIA now work together to operate important NVIDIA AI applications on the brand-new Oracle Cloud Infrastructure (OCI) SuperclusterTM. As the first hyperscale cloud service provider, NVIDIA has chosen OCI to provide NVIDIA DGX CloudTM, an AI supercomputing service, at a significant scale. Additionally, NVIDIA is operating NVIDIA AI Foundations on OCI, one of its brand-new generative AI cloud services that is accessible through DGX Cloud.
“The limitless opportunities for AI-driven innovation are helping transform virtually every business. NVIDIA’s collaboration with Oracle Cloud Infrastructure puts the extraordinary supercomputing performance of NVIDIA’s accelerated computing platform within reach of every enterprise,” said Manuvir Das, vice president of enterprise computing, NVIDIA.
New Supercluster for OCI
The OCI Supercluster, which was certified by NVIDIA to ensure it satisfies the stringent requirements of DGX Cloud, is used by NVIDIA DGX Cloud and the NVIDIA AI Foundations services that are run on it.
OCI Compute Bare Metal, an ultra-low latency RoCE cluster based on NVIDIA networking, and a selection of HPC storage are all included in OCI’s Supercluster. Numerous OCI Compute Bare Metal instances that can effectively handle massively parallel workloads have been deployed and validated by NVIDIA. The number of OCI Compute Bare Metal instances with 32,768 A100 GPUs that OCI Supercluster networking can support has increased to 4096. There are currently a few OCI Compute bare metal instances with NVIDIA H100 GPUs available. Additionally, NVIDIA said that Oracle will upgrade its networking stack with NVIDIA BlueField-3 DPUs.
AI Services for Generating Custom Business Models
The modeling services offered by NVIDIA AI Foundations include biology, language, pictures, video, and 3D. Businesses may create specialized, domain-specific, generative AI apps for intelligent conversation and customer assistance, professional content creation, digital simulation, and more using the NVIDIA NeMoTM language service and the NVIDIA Picasso image, video, and 3D services. The NVIDIA BioNeMoTM cloud service provides tools to swiftly configure and deploy generative AI applications for biology AI model training and inference.
On OCI, the OCI Supercluster, which contains purpose-designed RDMA networking that enables near line rate performance with microsecond latency and solves blocking difficulties for RDMA-dependent workloads, benefits custom models built with NVIDIA AI Foundations and model families like GPT-3.