Enterprise

NVIDIA Launches DSX Ready Qualification for AI Factory Components

New program certifies battery storage and cooling systems against reference design standards as infrastructure constraints reshape datacenter planning.

Omega Editorial· September 21, 2026· 3 min read

NVIDIA Introduces Component Qualification Program for AI Infrastructure

NVIDIA has launched DSX Ready, a qualification program designed to help AI datacenter builders identify power and cooling products that meet the company's DSX AI factory reference design requirements. The program addresses a growing challenge: as AI infrastructure scales, physical constraints around power delivery, cooling capacity, water availability and grid connections increasingly determine what computing systems builders can actually deploy.

The program initially covers two critical infrastructure categories—battery energy storage systems (BESS) and cooling distribution units (CDUs). NVIDIA plans to expand the qualification framework to additional infrastructure and software categories over time.

Why it matters

AI factory construction has shifted from a pure compute problem to a systems integration challenge where power and cooling often become the limiting factors. A qualification program that connects reference architectures to specific vendor products reduces integration risk and gives builders a clearer path from design to deployment. For infrastructure vendors, it provides a defined target for product development and a recognized credential for market differentiation.

Initial Qualified Partners

At launch, NVIDIA qualified BESS solutions from Hitachi Energy, LG Energy Solution and Tesla. For liquid cooling infrastructure, the company certified CDU offerings from LG Electronics, LiquidStack and Vertiv.

The qualification process varies by product category. BESS providers must run specified tests and submit supporting data for NVIDIA review within a defined qualification boundary. The program clarifies that passing qualification does not replace site-specific engineering or guarantee grid-level stability—those remain the builder's responsibility.

CDU providers use a self-qualification suite to demonstrate their products meet NVIDIA's functional requirements. Builders can then evaluate how qualified units fit their specific facility layouts and operating parameters.

Connecting Reference Designs to Real Products

The DSX Ready program extends NVIDIA's broader DSX AI factory platform, which treats compute, networking, power, cooling, facilities and software as a unified system. According to NVIDIA, this integrated approach helps operators produce more useful AI output within the physical constraints of their sites.

The qualification framework addresses a practical gap in the infrastructure market. Reference designs establish performance targets and integration patterns, but builders still need to identify which specific products meet those standards. DSX Ready creates a structured path from architecture to vendor selection.

NVIDIA notes that optimizing individual components can shift bottlenecks elsewhere in the system. A qualified cooling unit may meet thermal requirements while the builder still needs to verify water availability, facility space and maintenance access. The program aims to reduce uncertainty in the component selection phase while preserving the need for comprehensive site engineering.

Partners interested in qualifying products can connect with NVIDIA teams through the DSX Ready program page. The company has not disclosed qualification fees, testing timelines or the criteria for adding new product categories.

These details were first reported by NVIDIA in a company blog post announcing the program.

#nvidia#ai infrastructure#datacenter cooling#energy storage#ai factory

This is an original analysis by the Omega editorial team. Source reporting: AI Watch.

Want systems like this working for your business?

Book a Call

More in Enterprise

Enterprise· 3 min read

Positron IDE Now Runs on Amazon SageMaker for Unified Data Science

Posit's integrated development environment brings R and Python workflows together on managed infrastructure with governed data access and AI assistance.

Via AI Watch · Sep 21, 2026
Enterprise· 3 min read

Benchling Isolates AI Agent Code with Multi-Layer AWS Defense

Life sciences platform runs 600+ daily AI-generated code sessions across 250 tenants using DNS firewall, VPC endpoints, and per-job credentials.

Via AI Watch · Sep 21, 2026
Enterprise· 2 min read

Meta's Petal Cable to Deliver Petabit Capacity Across Atlantic

The 7,000-kilometer subsea link between France and the US will double transoceanic data capacity when it launches in 2029.

Via The Verge · Sep 21, 2026