Nvidia and Schneider Electric are testing a digital twin to rehearse AI data center upgrades before they go live.

**Nvidia and Schneider Electric are piloting a digital twin at Nscale's SIN01 data center in Portugal.** The system links AI workloads to models of the site's power and cooling, so operators can see what an infrastructure change would do before they make it. Shane Snider, a senior news writer at TechTarget, reported the pilot on Oct. 21, 2026.

Each partner contributes a piece. AVEVA supplies industrial engineering and operations software. ETAP models and simulates electrical power systems. Nscale is the AI infrastructure and cloud provider, and Start Campus is developing the data center campus in Portugal.

The twin is designed for a question that facilities engineers planning denser GPU deployments must eventually answer: can the existing electrical distribution and liquid cooling support the next generation of hardware? If not, the options are a costly redesign, a late go-live, or capacity left stranded. Stephen Sopko, who leads semiconductor and deep-tech research at HyperFrame Research, told TechTarget that the near-term payoff is lower commissioning risk and faster deployment. Efficiency gains, he said, come later, if they come at all.

**How the SIN01 twin is wired**

The system combines Schneider's energy management technology with AVEVA's industrial software, ETAP's electrical engineering tools and third-party technology. Facility engineering data and operational data feed into power and cooling simulations, which are tied to the requirements of the AI workloads. The practical use is testing higher rack density. An operator can model what a denser layout does to electrical distribution and liquid cooling before any rack is moved.

The companies say they want the system to cover the full AI factory lifecycle, from design through operations, and to make it easier to bring in equipment and models from different vendors. Sopko sees that interoperability as the most useful element, since it could let facility engineering tools work with AI infrastructure models rather than remaining in separate silos.

The concept is not new. On March 18, 2025, Facilities Dive's Joe Burns reported that Schneider and ETAP had released digital twins for designing and simulating the power needs of AI factories. Those twins were built on ETAP's integration of Nvidia Omniverse and drew on mechanical, thermal, networking and electrical inputs.

For teams writing specifications for future halls, this suggests that vendor twins built on this stack are still at the pilot stage rather than mature products. Specifications should be written with that in mind.

**What CoreWeave's DSX Air shows by comparison**

CoreWeave already takes a simulation-first approach. According to TechTarget, the company uses Nvidia's DSX Air, a cloud-based platform, to simulate AI factory environments and validate configurations before physical systems arrive. DSX Air models compute, networking, storage and software orchestration. Public disclosures do not show CoreWeave using it to control power and cooling in a live facility.

Side by side, the SIN01 pilot moves the rehearsal from the server stack into the building itself, covering switchgear-level electrical behavior and the cooling loops that keep dense racks within their limits. Sopko's point applies to both cases: rehearsing a configuration before hardware arrives can surface problems early, which could shorten validation and reduce physical rework.

The gap may be smaller than it appears, however. Sopko describes full bidirectional closed-loop control, in which the twin acts on the facility as well as reading from it, as a longer-term goal. That suggests the SIN01 twin is, for now, closer to a planning tool than a control system, and operators still make the decisions. The near-term payoff may therefore be smaller, with no immediate change in who is responsible on the floor.

**Sized for Rubin-class density**

The campus figures explain why a twin is worth building here. The SINES Data Campus that SIN01 belongs to has a design capacity of 1.2 GW and is expected to house more than 66,000 Nvidia Rubin GPUs, though no deployment schedule has been given. TechTarget reports that the companies have not said how many Rubin GPUs are currently in place at SIN01 or when the rest will arrive. A build-out of this size, done in phases, is where a model of the next phase earns its keep: every hall added is a chance to find a power or cooling mismatch in software instead of on site.

Schneider's cooling hardware fits the same design pattern. Motivair by Schneider Electric's MCDU-70 coolant distribution unit is rated at 2.5 MW, and six of them can provide 4+2 redundancy for a 10 MW DSX Blueprint deployment, according to figures reported by TechTarget and Motivair by Schneider Electric via North American Clean Energy.

Nothing in the reporting says SIN01 uses the MCDU-70. What the pairing suggests is that the 10 MW blueprint block is becoming a shared unit of planning, with cooling hardware specified around it. This matters most for operators buying into the Nvidia reference design. Teams with mixed vendors or non-reference halls would benefit more from the multi-vendor interoperability the companies describe.

**The scorecard SIN01 still has to post**

Sopko set out a specific test for whether a twin improves AI computing efficiency. Compare time to first token, commissioning defects, power usage effectiveness, thermal events and measured compute output against a baseline using comparable workloads and hardware. Simulation results alone do not count. The baseline is what turns a promising model into evidence.

The companies have not yet published SIN01 results for token throughput, energy use, cooling efficiency or deployment time. They have also not released validation comparing the twin's predictions with measured operating conditions, which is the comparison that shows whether the model can be trusted.

As Sopko put it, the nearer-term value is lower commissioning risk and faster deployment. Efficiency gains still have to be demonstrated against a baseline using comparable workloads and hardware.

Before a twin goes into a capital plan, ask the vendor for one site where predicted power and thermal behavior was checked against measured conditions, and for the baseline used to compare commissioning defects.

The next things to watch at SIN01 are that prediction-versus-measurement comparison and a Rubin deployment schedule, neither of which the companies have yet provided. Fewer commissioning defects measured against a comparable baseline would be one sign that the twin is doing its job.