NVIDIA and Microsoft Collaborate to Build a World-Class AI Supercomputer

In a recent collaboration, NVIDIA and Microsoft are working together to build one of the world's most powerful AI supercomputers, which will be powered by Microsoft Azure's advanced supercomputing infrastructure along with NVIDIA GPUs, networking, and AI software.

This new AI supercomputer will help enterprises train, deploy, and scale AI, including large, state-of-the-art models. Azure's AI-optimized virtual machine instances will be the first public cloud to incorporate NVIDIA's advanced AI stack, including NVIDIA's A100 and H100 GPUs, Quantum-2 400Gb/s InfiniBand networking, and AI Enterprise software suite.

“AI technology advances as well as industry adoption are accelerating. The breakthrough of foundation models has triggered a tidal wave of research, fostered new startups and enabled new enterprise applications,” said Manuvir Das, vice president of enterprise computing at NVIDIA. “Our collaboration with Microsoft will provide researchers and companies with state-of-the-art AI infrastructure and software to capitalize on the transformative power of AI.”

“AI is fueling the next wave of automation across enterprises and industrial computing, enabling organizations to do more with less as they navigate economic uncertainties,” said Scott Guthrie, executive vice president of the Cloud + AI Group at Microsoft. “Our collaboration with NVIDIA unlocks the world’s most scalable supercomputer platform, which delivers state-of-the-art AI capabilities for every enterprise on Microsoft Azure.”

The collaboration will also focus on optimizing Microsoft's DeepSpeed deep learning optimization software and making NVIDIA's full stack of AI workflows and software development kits optimized for Azure available to Azure enterprise customers. The platform will support a wide range of AI applications and services, including Microsoft DeepSpeed and NVIDIA AI Enterprise. NVIDIA AI Enterprise streamlines the entire AI workflow, from data processing to AI model training, simulation, and large-scale deployment.

With the combination of NVIDIA Compute and Quantum-2 InfiniBand on Azure, customers will be able to train even the largest language models and build complex recommender systems at scale. The AI-optimized virtual machine instances will provide scalable peak performance for AI training and deep learning inference workloads of any size.

This collaboration between NVIDIA and Microsoft is a significant step towards advancing AI development and deployment, unlocking the world's most scalable supercomputer platform, and delivering state-of-the-art AI capabilities for every enterprise on Microsoft Azure.