Microsoft CEO Satya Nadella announced that the company's Azure cloud platform has become the first cloud service provider to begin validation of NVIDIA's Vera Rubin NVL72 system. Nadella stated in a social media post on Friday afternoon: "We are the first cloud platform to bring NVIDIA's Vera Rubin NVL72 system into validation, marking another significant leap in our joint effort with NVIDIA to build next-generation AI infrastructure." The long-standing partnership between Microsoft and NVIDIA enables better deployment of continuously iterated hardware equipment within Microsoft's data centers. Rani Borkar, President of Microsoft Azure Hardware Systems, commented, "Microsoft possesses years of market-proven experience in designing and deploying scalable AI infrastructure that evolves with each major advancement in AI technology." Borkar added, "Synchronizing with each generation of NVIDIA's accelerated computing infrastructure, Microsoft rapidly integrates NVIDIA's innovations and achieves scalable delivery. The NVIDIA Rubin platform represents an important step forward in accelerated computing, and Azure's AI data centers and supercomputing factories were designed from the outset to fully leverage this advantage. Over the years, we have co-designed with NVIDIA in areas including interconnects, memory systems, thermal management, packaging, and rack-level architecture, meaning Rubin can be directly integrated into the Azure platform without requiring redesign." Although Microsoft claims to be the first company to introduce Vera Rubin for validation, multiple cloud providers are expected to deploy the technology by 2026. These include Amazon AWS, CoreWeave, Google, Nebius, and Oracle. According to NVIDIA's data, each Vera Rubin NVL72 rack can deliver up to 3.6 exaflops of performance, approximately five times the performance of GB200-based systems. Each rack connects 72 GPUs and 36 CPUs through sixth-generation NVLink, providing 260 TB/s of bandwidth.
Comments