Microsoft Azure Unveils World's First NVIDIA GB300 NVL72 Supercomputing Cluster for AI Workloads

Reviewed byNidhi Govil

6 Sources

Share

Microsoft has deployed a groundbreaking supercomputer-scale cluster of NVIDIA GB300 NVL72 systems on its Azure platform, featuring over 4,600 Blackwell Ultra GPUs. This state-of-the-art infrastructure aims to accelerate AI model training and inference for OpenAI's most demanding workloads.

Microsoft Azure's Groundbreaking AI Infrastructure Upgrade

Microsoft has unveiled a revolutionary upgrade to its Azure cloud platform, deploying the world's first large-scale NVIDIA GB300 NVL72 supercomputing cluster

1

2

. This cutting-edge infrastructure, featuring over 4,600 NVIDIA Blackwell Ultra GPUs, is set to redefine the landscape of AI model training and inference

3

.

Source: NDTV Gadgets 360

Source: NDTV Gadgets 360

Technical Specifications and Performance

The new Azure NDv6 GB300 VM series boasts impressive specifications:

  • 72 NVIDIA Blackwell Ultra GPUs and 36 NVIDIA Grace CPUs per rack

    3

  • 800 Gb/s per GPU cross-rack bandwidth via NVIDIA Quantum-X800 InfiniBand

    3

  • 130 TB/s of NVIDIA NVLink bandwidth within each rack

    2

  • 37 TB of fast memory

    2

  • Up to 1,440 petaflops of FP4 Tensor Core performance

    3

This powerful configuration allows the cluster to function as a single, unified accelerator, capable of handling massive AI workloads with unprecedented efficiency

1

.

Source: NVIDIA Blog

Source: NVIDIA Blog

Impact on AI Development

Microsoft claims that this new infrastructure can dramatically reduce AI model training times from months to weeks

3

. The cluster is optimized for reasoning models, agentic AI systems, and multimodal generative AI workloads, supporting the training of models with hundreds of trillions of parameters

4

.

Collaboration and Future Prospects

The deployment of this cluster is the result of a deep partnership between NVIDIA and Microsoft, aimed at building AI infrastructure for the world's most demanding workloads

2

. It's particularly noteworthy that this cluster will be dedicated to OpenAI workloads, furthering the collaboration between these tech giants

1

.

Technological Innovations

To achieve this level of performance, Microsoft had to reimagine every layer of its data center

2

. Key innovations include:

  • Custom liquid cooling systems using standalone heat exchangers and facility loops

    1

  • Advanced power distribution models

    5

  • Reengineered software stack for orchestration and storage

    4

Industry Implications

This deployment marks a significant milestone in the advancement of AI infrastructure. Ian Buck, VP of Hyperscale and High-performance Computing at NVIDIA, stated that this system "sets the definitive new standard for accelerated computing"

3

. As Microsoft aims to scale to hundreds of thousands of NVIDIA Blackwell Ultra GPUs globally, it's poised to unlock new frontiers in AI research and development

5

.

Conclusion

The introduction of Microsoft's NVIDIA GB300 NVL72 supercomputing cluster represents a leap forward in AI infrastructure. As this technology scales and evolves, it promises to accelerate AI innovation across various sectors, potentially leading to breakthroughs in fields ranging from scientific research to commercial applications.

Today's Top Stories

TheOutpost.ai

Your Daily Dose of Curated AI News

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

© 2025 Triveous Technologies Private Limited
Instagram logo
LinkedIn logo