Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore Azure's cutting-edge AI infrastructure developments in this conference talk from Microsoft Ignite 2025, where speakers Param Shah and Matt Vegas demonstrate how Microsoft is scaling cloud infrastructure to meet the demands of frontier and enterprise AI workloads. Discover Azure's comprehensive silicon-to-systems optimization approach, including hardware-software codesign principles and revolutionary datacenter advances in cooling, power management, networking, and security protocols. Learn about the latest AI infrastructure powered by NVIDIA Grace Blackwell Superchips and Quantum-2 InfiniBand technology, featuring the new ND GB200/GB300 virtual machines that deliver exascale performance capabilities and achieve over 860,000 tokens per second on LLAMA 70B models. Examine the NC H100 and NC RTX Blackwell virtual machines specifically designed for enterprise inferencing workloads, understanding how these solutions address the growing computational requirements of modern AI applications. Gain insights into Microsoft's strategic approach to building scalable AI infrastructure that bridges the gap between cutting-edge research capabilities and practical enterprise deployment needs, with detailed technical demonstrations and real-world performance metrics that showcase the potential of next-generation cloud computing for AI workloads.
