Key points
- Microsoft Azure has launched the first at-scale production cluster with over 4,600 NVIDIA GB300 NVL72, featuring NVIDIA Blackwell Ultra GPUs connected through the next-generation NVIDIA InfiniBand network.
- This cluster is the first of many, as Microsoft scales to hundreds of thousands of Blackwell Ultra GPUs deployed across its AI datacenters globally, enabling model training in weeks instead of months.
- The ND GB300 v6 VMs are optimized for reasoning models, agentic AI systems, and multimodal generative AI, and are poised to become the new standard for AI infrastructure.
According to sources, Microsoft Azure has made a significant breakthrough in the field of artificial intelligence with the launch of the NVIDIA GB300 NVL72 supercluster. This co-engineered system delivers the world’s first at-scale GB300 production cluster, providing the supercomputing engine needed for OpenAI to serve multitrillion-parameter models. This sets the definitive new standard for accelerated computing, as stated by Ian Buck, Vice President of Hyperscale and High-performance Computing at NVIDIA.
The ND GB300 v6 VMs are the latest iteration of Azure’s virtual machines, accelerated by NVIDIA’s Blackwell architecture. These VMs are optimized for reasoning models, agentic AI systems, and multimodal generative AI, and are built on a rack-scale system with 18 VMs and a total of 72 NVIDIA Blackwell Ultra GPUs. Each rack has a total of 37TB of fast memory and delivers up to 1,440 petaflops of FP4 Tensor Core performance.
Microsoft Azure has been working closely with NVIDIA to build infrastructure for frontier AI, which requires reimagining every layer of the stack, including computing, memory, networking, datacenters, cooling, and power. The ND GB300 v6 VMs are a clear representation of this transformation, with NVLink and NVSwitch reducing memory and bandwidth constraints, enabling up to 130TB per second of intra-rack data-transfer.
As Microsoft Azure continues to ramp up GB300 worldwide deployments, customers can expect to train and deploy new models in a fraction of the time compared to previous generations. The ND GB300 v6 VMs are poised to become the new standard for AI infrastructure, and Azure is proud to lead the way, supporting customers to advance frontier AI development. With its advanced cooling systems, custom protocols, and in-network computing, Azure is well-positioned to deliver unprecedented levels of performance at high efficiency to its customers. Microsoft has invested heavily in AI infrastructure over the years, and this latest development is a testament to its commitment to redefining AI infrastructure and collaboration with NVIDIA.
Read the rest: Source Link
You might also like: Why Choose Azure Managed Applications for Your Business & How to download Azure Data Studio.
Remember to like our facebook and our twitter @WindowsMode for a chance to win a free Surface every month.
Discover more from Windows Mode
Subscribe to get the latest posts sent to your email.