← Back
NVIDIA Blog

OCP Global Summit: Amid Ecosystem Adoption, NVIDIA Opens Vera Rubin Architecture to Help Scale AI Factories Worldwide

•11 min read•
#compute#deployment#nvidia
OCP Global Summit: Amid Ecosystem Adoption, NVIDIA Opens Vera Rubin Architecture to Help Scale AI Factories Worldwide
Level:Advanced
For:AI Engineers
✦TL;DR

NVIDIA announced the MG X 1.2 open‑architecture update at the OCP Global Summit, extending rack‑scale AI factory designs to 250 kW deployments with blind‑mate liquid cooling, a 5,000 A busbar, and native 19‑inch equipment support. New open standards include the Ultra‑Dense Small Form‑Factor Pluggable (USFP) 400 Gb/s networking module, Spectrum‑X Ethernet SuperNIC, Spectrum‑6 innovations, and Falcon Transport on future ConnectX‑10 SuperNICs. The contributions also add 800 V DC AI‑factory designs from Visionbay and open‑source software components for NVIDIA DSX AI Factory, aiming to accelerate deployment, improve power efficiency, and lower build‑out time for large‑scale AI infrastructure. Engineers can leverage these standards to design interoperable, high‑density AI racks that simplify manufacturing, service, and power‑grid integration.

⚡ Key Takeaways

  • MG X 1.2 specification supports rack deployments up to 250 kW and includes a liquid‑cooled 5,000 A busbar.
  • Introduces the Ultra‑Dense Small Form‑Factor Pluggable (USFP) standard, the industry’s first 400 Gb/s networking fabric module.
  • Adds blind‑mate liquid‑cooling, configurable I/O bays, fanless liquid‑cooled compute trays, and native 19‑inch equipment for higher‑density racks.
  • Enables Falcon Transport support on future NVIDIA ConnectX‑10 SuperNICs and BlueField DPUs.
  • Visionbay’s 800 V DC AI‑factory design is announced for the Vera Rubin era, extending power‑grid resilience.
💡 Why It Matters

By adopting the MG X 1.2 open standards, data‑center teams can rapidly build gigawatt‑scale AI factories with standardized, high‑density racks that reduce power‑cost per token and cut engineering time for assembly and servicing.

✅ Practical Steps

  1. Integrate the USFP 400 Gb/s module and Spectrum‑X/Spectrum‑6 SuperNICs into new rack designs following the MG X 1.2 specifications.
  2. Use the open‑source NVIDIA DSX AI Factory software components to manage liquid‑cooling and power‑busbar interfaces as defined in the MG X 1.2 reference design.

Want the full story? Read the original article.

Read on NVIDIA Blog ↗

More like this

New in AWS Quick: More capable flows, an updated web experience, and agents that work together

AWS ML Blog•#agents

Three insights you may have missed from theCUBE’s coverage of the Dell Technologies New York Symposium

SiliconANGLE AI•#deployment

Welcome Inkling 2 Mini, Cutting-edge Performance for your Local Setup

Hugging Face Blog•#llm

NVIDIA and Microsoft Expand Accelerated Computing From Azure to Windows PCs

NVIDIA Blog•#compute

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING