1 min read
Add as a preferred source on Google

Nvidia Says DSX POD Could Cut AI Factory Deployment to 14 Days

The Vera Rubin reference design moves integration and testing into modular containers, with partners planning worldwide deliveries in early 2027.

A modular computing container stands in an industrial assembly bay / TokenPost.ai
A modular computing container stands in an industrial assembly bay / TokenPost.ai

Nvidia introduced DSX POD at GTC Berlin on Oct. 21, a reference design for Vera Rubin AI factories aimed at shortening deployment by moving integration and testing into factory-built modules.

At prepared sites, the design is intended to reduce deployment time from six months for a conventional build to as little as 14 days. DSX POD packages computing equipment, networking, rack power and liquid cooling in modular containers. Partners assemble and test the systems before shipment, while site crews can prepare power, cooling and network connections at the same time.

After delivery, crews still need to connect the containers to site infrastructure and commission the systems before they are operational.

“An AI factory creates value when it starts producing tokens — not when construction is complete,” Nvidia founder and CEO Jensen Huang said.

Dell Technologies, HPE, Lenovo, Schneider Electric and Vertiv are among the partners planning worldwide deliveries of DSX POD systems in early 2027.

Simon Yoon

Reporter

Simon Yoon reports on blockchain technology for TokenPost. Send corrections or tips to info@tokenpost.com.

Loading…