Microsoft and Nvidia have unveiled groundbreaking solutions merging Nvidia’s generative AI and Omniverse technologies with the robust capabilities of Microsoft Azure services.
Rani Borkar, Corporate Vice President of Azure Hardware Systems and Infrastructure at Microsoft, announced the integration of Nvidia’s Grace Blackwell 200 (GB200) Superchip with Microsoft Azure. Designed for large-scale generative AI workloads and high-performance computing, the GB200 boasts an astounding 16 TBs of memory bandwidth and up to 45 times the inference power on trillion-parameter models compared to its predecessors.
Microsoft plans to introduce GB200-based instances on Azure, empowering customers to develop and deploy cutting-edge AI solutions capable of handling vast data and complexities, thereby accelerating time-to-market.
Additionally, Microsoft has launched the Azure NC H100 v5 VM series, based on Nvidia’s H100 NVL platform, allowing for GPU partitioning into up to seven instances. This flexibility ensures scalability across diverse AI workloads.
Jensen Huang, Founder and CEO of Nvidia, emphasized the transformative potential of AI in everyday life, highlighting the collaboration with Microsoft as pivotal in realizing this vision. The partnership aims to unlock AI’s promise for customers, fostering innovation and delivering solutions that shape the future.
Furthermore, the integration of Azure with Nvidia DGX Cloud and the Nvidia Clara suite targets enhanced efficiency in healthcare and life sciences. Later this year, manufacturers will gain access to Nvidia Omniverse Cloud APIs on Azure, facilitating increased data interoperability, collaboration, and physics-based visualization within existing software applications.
Satya Nadella, Chairman and CEO of Microsoft, underscored the joint efforts to make AI advancements tangible, driving benefits and productivity gains globally. From bringing the GB200 processor to Azure to enabling seamless integrations between DGX Cloud and Microsoft Fabric, the collaboration offers comprehensive platforms and tools across every layer of the AI stack, empowering organizations to unleash breakthrough capabilities.