Presented by Microsoft
Microsoft’s announcements about brand-new collaborations with long-standing partner Nvidia put the company at the forefront of this year’s Nvdia GTC AI conference in San Jose, March 18 – 21.
The week’s round of AI innovation news ran the gamut from AI infrastructure and service advances to new platform integrations, industry breakthroughs and more. Plus, Nidhi Chappell,V P of Azure Generative AI and HPC Platform Microsoft, sat down for an exclusive one-on-one conversation with VentureBeat Senior Writer Sharon Goldman to talk about Microsoft’s partnership with both OpenAI and Nvidia, where the market is headed and more.
“If you look at what got us to here, partnership is really at the center of everything we do. When you’re training a large foundational model, you want to have infrastructure at large scale that can run for a long period of time,” Chappell said. “We’ve invested a lot of time and effort with Nvidia to make sure we can deliver performance, we can do it reliably, and we can do it globally across the world so that [using our Azure OpenAI service] enterprise customers can seamlessly integrate that in their existing flows or they can start their new work on our tool.”
Watch the full interview below, Live from GTC: A Conversation with Microsoft | NVIDIA On-Demand, read on for a look at the major conference announcements and don’t miss Microsoft’s in-depth series of panels and talks, all free to watch on demand.
AI infrastructure levels up with major new integrations
Workloads are getting more sophisticated and requiring more heavy lifting – which means hardware innovation has to step in. Announcements to that end: first, Microsoft is one of the first organizations to use the Nvidia G200 Grace Blackwell Superchip and Nvidia Quantum-X800 InfiniBand networking, integrating these into Azure. Plus, the Azure NC H100 v5 VM virtual machine series is now available to organizations of every size.
The Nvidia G200 Grace Blackwell Superchip is specifically designed to handle the heavy lifting of increasingly complex AI workloads, high-performing workloads and data processing. New Azure instances based on the latest GB200 and recently announced Nvidia Quantum-X800 InfiniBand networking will help accelerate frontier and foundational models for natural language processing, computer vision, speech recognition and more. It features up to 16 TB/s of memory bandwidth and up to an estimated 45 times greater inference on trillion parameter models than the previous generation. The Nvidia Quantum-X800 InfiniBand networking platform works to extend the GB200’s parallel computing tasks into massive GPU scale.
Learn more about the Nvidia and Microsoft integrations here.
The Azure NC H100 v5 VM series, built for mid-range training, inferencing and high-performance compute (HPC) simulations, is now available to organizations of every size. The VM series is based on the Nvidia H100 NVL platform, which is available with one or two Nvidia H100 94GB PCIe Tensor Core GPUs connected by NVLink with 600 GB/s of bandwidth.
It supports 128GB/s bi-directional communication between the host processor and the GPU to reduce data transfer latency and overhead to make AI and HPC applications faster and more scalable. With Nvidia multi-instance GPU (MIG) technology support, customers can also partition each GPU into up to seven instances.
See what customers are achieving now.
Major breakthroughs in healthcare and life sciences
AI has been a major breakthrough for rapid-paced innovations in medicine and the life sciences, from research to drug discovery and patient care. The expanded collaboration pairs Microsoft Azure with Nvidia DGX Cloud and the Nvidia Clara suite of microservices to give healthcare providers, pharmaceutical and biotechnology companies and medical device developers the ability to fast track innovation in clinical research, drug discovery and patient care.
The list of organizations already leveraging cloud computing and AI include: Sanofi, the Broad Institute of MIT and Harvard, Flywheel and Sophia Genetics, academic medical centers like the University of Wisconsin School of Medicine and Public Health, and health systems like Mass General Brigham. They’re driving transformative changes in healthcare, enhancing patient care and democratizing AI for healthcare professionals and more.
Learn how AI is transforming the healthcare industry.
Industrial digital twins gaining traction with Omniverse APIs on Azure
Nvidia Omniverse Cloud APIs are coming to Microsoft Azure, extending the Omniverse platform’s reach. Developers can now integrate core Omniverse technologies directly into existing design and automation software applications for digital twins, or their simulation workflows for testing and validating autonomous machines like robots or self-driving vehicles.
Microsoft demonstrated a preview of what’s possible using Omniverse Cloud APIs on Azure. For instance, factory operators can see real-time factory data overlaid on a 3D digital twin of their facility to gain new insights that can speed up production.
In his GTC keynote, Nvidia CEO Jensen Huang showed Teamcenter X connected to Omniverse APIs, giving the software the ability to connect design data to Nvidia generative AI APIs, and use Omniverse RTX rendering directly inside the app.
Learn more about the ways organizations are deploying Omniverse Cloud APIs in Azure.
Enhancing real-time contextualized intelligence
Copilot for Microsoft 365, soon available as a dedicated physical keyboard key on Windows 11 PCs, combines the power of large language models with proprietary enterprise data. Nvidia GPUs and Nvidia Triton Inference Server power up AI inference predictions for real-time intelligence that’s contextualized, enabling users to enhance their creativity, productivity and skills.
Turbocharging AI training and AI deployment
Nvidia NIM inference microservices, part of the Nvidia AI Enterprise software platform, provides cloud-native microservices for optimized inference on more than two dozen popular foundation models. For deployment, the microservices deliver prebuilt, run-anywhere containers powered by Nvidia AI Enterprise inference software — including Triton Inference Server, TensorRT and TensorRT-LLM — to help developers speed time to market of performance-optimized production AI applications.
Integration of Nvidia DGX Cloud with Microsoft Fabric gets deeper
Microsoft and Nvidia are pairing up to ensure Microsoft Fabric, the all-in-one analytics solution for enterprises, is further integrated into Nvidia DGX Cloud compute. That means that Nvidia’s workload-specific optimized runtimes, LLMs and machine learning will work seamlessly with Microsoft Fabric. With Fabric OneLake as the underlying data storage, developers can apply data-intensive use cases like digital twins and weather forecasting. The integration also gives customers the option to use DGX Cloud to accelerate their Fabric data science and data engineering workloads.
See what you missed at GTC 2024
Microsoft dove into the powerful potential of all its collaborations with Nvidia, and demonstrated why Azure is a critical component of a successful AI strategy for organizations at every size. Watch all of Microsoft’s panels and talks here, free to stream on demand.
Learn more about Microsoft and NVIDIA AI solutions:
VB Lab Insights content is created in collaboration with a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. For more information, contact [email protected].
The post Microsoft and Nvidia announce major new integrations, breakthroughs and more at GTC appeared first on Venture Beat.
