At a glance
- Companies around the world are building on the NVIDIA DSX AI Factory Platform to plan, simulate and run the grid and the AI factory as one system, from the substation to the rack
- Their adoption of the platform responds to an acute problem
- Power is now the biggest bottleneck in infrastructure
- More than $300 billion…
# How Companies Are Using NVIDIA DSX to Break the Power Bottleneck
The rapid expansion of AI compute is colliding with a hard physical limit: the electrical grid. Across the industry, power has emerged as the primary constraint on data center growth, leaving more than $300 billion in equipment backlogged and over 200 gigawatts of projects stalled in U.S. interconnection queues. NVIDIA’s DSX AI Factory Platform is designed to address this by treating the grid, the facility, and the compute layer as a single integrated system rather than isolated silos.
The Cost of Siloed Infrastructure
Historically, grid infrastructure, facility design, and compute deployment have been planned independently. This fragmentation creates delays, increases operational risk, and leaves capacity stranded. For operators building AI factories, the inability to synchronize power availability with compute demand means projects can be ready for hardware but lack the electrical capacity to run it.
NVIDIA DSX aims to close this gap by unifying simulation, demand management, and hardware-level power optimization across the entire stack—from the substation to the rack.
Simulation Before Groundbreaking
A core component of the platform is NVIDIA DSX Sim, which pairs with SimReady assets from leading energy manufacturers. Engineers can model the grid, substation, and AI factory together, validating generators, switchgear, power systems, and cooling before construction begins. This pre-build validation is intended to reduce the risk of costly redesigns and ensure that power infrastructure aligns with actual compute loads.
The simulation layer extends into operations through NVIDIA Omniverse-powered digital twins. Schneider Electric, working with AVEVA and ETAP, is advancing DSX with validated NVIDIA Vera Rubin power and cooling reference architectures. By combining AVEVA’s engineering software with NVIDIA SimReady assets and OpenUSD, the collaboration connects simulations, real-time operational data, and AI-driven optimization across Schneider Electric, AVEVA, and third-party solutions.
Turning AI Factories Into Grid Assets
Beyond planning, DSX Flex addresses how AI factories interact with live grid conditions. The capability syncs AI factory demand with grid status and onsite generation, effectively converting data centers from static loads into flexible grid assets. This flexibility is increasingly valuable as utilities manage constrained transmission and distribution networks.
Phaidra has integrated DSX Flex, DSX Max-Q, and DSX Exchange into its self-learning agents. The company reports optimizing cooling and power allocation to maximize tokens per watt, improve grid flexibility, and deliver approximately 10% more compute within the same power envelope.
Emerald AI has taken a different approach, integrating its Conductor platform into DSX Flex to help utilities identify underused grid capacity. The approach is modeled to unlock up to 100 gigawatts on the existing U.S. grid without new transmission buildout.
Squeezing Compute From Every Watt
At the hardware level, NVIDIA DSX MaxLPS and DSX Exchange are designed to increase compute efficiency per watt and enable secure signal sharing across compute, networking, energy, cooling, and operations. These capabilities feed into the broader goal of improving energy efficiency and workload performance while accelerating time to token.
Partner Ecosystem Aligns With Vera Rubin
Several major energy and infrastructure vendors are aligning their product roadmaps with the NVIDIA DSX reference architecture, particularly around the NVIDIA Vera Rubin AI factory design.
GE Vernova is deploying its grid-to-facility and rack power expertise to complement DSX systems, with the goal of helping customers transform power constraints into competitive advantages while maximizing compute and token output, sustainability, and grid reliability.
Hitachi Energy’s battery energy storage system has qualified as NVIDIA DSX Ready, helping customers identify power infrastructure that meets applicable DSX reference architecture design requirements for resilient, reliable, and scalable AI factories.
Siemens Energy is expanding its collaboration with NVIDIA by developing a reference design for resilient, scalable, and grid-aware power infrastructure across generation, transmission, and distribution. The design is aligned with the requirements of the NVIDIA DSX Vera Rubin AI factory design. Siemens Energy is also evaluating NVIDIA technologies such as RAPIDS, Metropolis, and Isaac Sim for its Noedra digital twin platform, with potential applications in grid infrastructure monitoring, emerging risk identification, and unplanned outage prevention.
Operator Implications
For data center operators and AI infrastructure planners, the DSX platform signals a shift in how power is managed. Rather than treating electrical capacity as a fixed external constraint, operators can simulate, optimize, and dynamically adjust power use in coordination with grid conditions. The promise is faster deployment, higher utilization of existing infrastructure, and reduced risk of stranded assets.
NVIDIA is showcasing the Omniverse DSX Blueprint and the Vera Rubin DSX AI Factory Reference Design, with further details expected from NVIDIA and energy leaders at GTC Berlin. Schneider Electric will be at booth 3041 in hall 3, Hitachi Energy at booth 7004 in hall 7, and Siemens Energy at booth 3048 in hall 3. Schneider Electric sessions are scheduled for Wednesday, Oct. 21, including “From Grid to AI: Engineering the Future of AI Factories” and “Energy Intelligence: Powering the AI Era.”
MegaWatt Intelligence Desk
Hyperscale & Grid Analyst. `More stories by $MegaWatt Intelligence Desk →`
