Tag: Graphics processing unit

AI AnalyticsSemiconductors

AMD Gains Oracle Partnership in AI Chip Market Expansion

Oracle will deploy 50,000 AMD Instinct MI450 GPUs across its data centers starting 2026. This partnership follows AMD’s recent multi-billion dollar OpenAI agreement, signaling growing competition in the AI chip sector against market leader Nvidia.

In a significant development for the AI chip wars, AMD has secured another major partnership, this time with enterprise cloud giant Oracle Corporation, marking continued momentum for the chipmaker against dominant player Nvidia. The companies announced Tuesday that AMD will deploy 50,000 of its most advanced artificial intelligence chips within Oracle’s cloud infrastructure beginning in the second half of 2026.

Oracle-AMD AI Infrastructure Partnership Details

by Natalie Brooks
AI AnalyticsSemiconductors

U.S. AI Chip Strategy Intensifies As China Embraces Open Source Models

The global AI race reaches new intensity as U.S. tech giants form trillion-dollar chip partnerships while China pivots to open source models. This strategic divergence reshapes the future of artificial intelligence development and deployment worldwide.

The global AI race has escalated dramatically as United States technology giants forge unprecedented partnerships to control the entire AI production stack, while China’s leading firms embrace open source artificial intelligence models and domestic chip optimization. This strategic divergence represents a fundamental shift in how nations approach AI supremacy, with the U.S. building vertically integrated supply chains and China prioritizing adaptability through collaborative development.

U.S. Chip Ecosystem Forms Trillion-Dollar Loop

by Darren Holt
AI AnalyticsInfrastructure

Rack-Level Containment: The Future of AI Infrastructure Cooling

As AI workloads push rack densities beyond 100kW, traditional cooling methods are failing. Rack-level containment emerges as the critical solution, enabling ultra-high-density cooling without facility redesigns. This approach supports hybrid architectures and accelerates AI deployment.

Rack-level containment is rapidly becoming the cornerstone of modern AI infrastructure, addressing the unprecedented thermal challenges posed by generative AI workloads. As data centers confront rack densities exceeding 100kW and heading toward 1MW, traditional air cooling systems are proving inadequate. This evolution represents a fundamental shift in how we approach data center design, moving from room-level to rack-level thermal management strategies that can support tomorrow’s AI supercomputers starting today.

The AI Density Revolution Transforming Data Centers

by Darren Holt