NVIDIA announced this Wednesday, November 16, a multi-year collaboration with Microsoft to build an AI supercomputer that is expected to be among the most powerful in the world, powered by Microsoft Azure’s advanced supercomputing infrastructure combined with NVIDIA GPUs, the solution network and full stack of AI software to help companies train, deploy and scale AI, including large edge models.
Azure’s cloud-based AI supercomputer will include powerful and scalable ND and NC-series virtual machines optimized for distributed AI training and inference. It’s the first public cloud to integrate NVIDIA’s advanced AI stack, adding tens of thousands of NVIDIA A100 and H100 GPUs, NVIDIA Quantum-2400Gb/s InfiniBand networking, and NVIDIA AI Enterprise software suite to its platform.
Current Azure instances offer previous-generation Nvidia A100 GPUs paired with Quantum 200 Gb/s InfiniBand networking. Combined with Azure’s advanced compute cloud infrastructure, networking and storage, the new instances will deliver scalable peak performance for AI training and inference workloads. deep learning of any size.
NVIDIA AI Enterprise, which includes NVIDIA Riva for voice AI and NVIDIA Morpheus cybersecurity application frameworks, will streamline every step of the AI workflow, from data processing and AI model training to simulation and large-scale deployment.
As part of this collaboration, NVIDIA will use Azure’s scalable virtual machine instances to research and accelerate advancements in generative AI, an emerging field of AI in which foundational designs such as the Megatron Turing NLG 530B, ( the 530 billion parameter language model presented in October 2021 by Microsoft and NVIDIA), form the basis of unsupervised self-learning algorithms to create new text, code, digital images, video or audio.
The companies will also collaborate to optimize Microsoft’s DeepSpeed software suite. NVIDIA’s full stack of AI workflows and SDKs, optimized for Azure, will be made available to enterprise Azure customers.
Manuvir Das, Vice President of Enterprise Computing at NVIDIA, says:
“AI technology advancements and industry adoption are accelerating. The breakthrough in foundation models has triggered a tidal wave of research, fostered new startups and enabled new enterprise applications. Our collaboration with Microsoft will provide researchers and businesses with state-of-the-art AI infrastructure and software to harness the transformative power of AI. »
Scott Guthrie, Executive Vice President of Cloud + AI Group at Microsoft, concludes:
“AI is powering the next wave of automation in business and industrial computing, enabling organizations to do more with less as they navigate economic uncertainties. Our collaboration with NVIDIA unlocks the world’s most scalable supercomputing platform, delivering industry-leading AI capabilities for every business on Microsoft Azure. »