New multi‑source architecture designed to unlock high‑density AI data centers in grid‑constrained locations worldwide
Modular data center pioneer ECL today announced ECL FlexGrid™, a new power‑agnostic platform designed to unlock a global wave of AI at scale from massive training clusters to the inferencing edge by turning power‑constrained sites into high‑density, inference‑ready facilities. FlexGrid enables operators to deploy GPU‑rich capacity anywhere, including close to users and data sources, even where grid power and hydrogen are not readily available.
FlexGrid leverages ECL’s proprietary power conditioning system to intelligently combine multiple sources of energy ranging from grid power to hydrogen, natural gas, renewables and even diesel into a single, reliable AC or DC power feed for the data center. This allows customers to start with modest grid connections in the 2–10 MW range and scale to 20–25 MW of available capacity per edge site by layering in additional sources of power.
“The money, the growth and the real infrastructure challenges in AI are rapidly expanding from training to inferencing, and creating strong value and opportunity as a result,” said Yuval Bachar, founder and CEO of ECL. “Inference has to live close to people, data and applications, in and around major cities, smaller metros and industrial hubs where there is rarely a spare 50 or 100 megawatts sitting on the grid, and almost never a mature hydrogen ecosystem. FlexGrid was built exactly for the market conditions we face today. We take whatever energy source is available locally, normalize it through our power conditioning system, and deliver clean, reliable power to AI data centers at the edge.”
Unlike traditional facilities engineered around a single dominant energy source, FlexGrid was architected from day one to be fuel‑agnostic, so adding natural gas turbines, hydrogen systems or other prime movers does not require re‑designing the data center. This flexibility is critical as policy priorities and energy markets evolve.
“The industry has spent the last few years focused on building massive training factories, but the next decade will be defined by how well we deploy and power inferencing everywhere people live and work,” said Dean Nelson, CEO of Cato Digital, and founder and chairman of Infrastructure Masons (iMasons). “What makes ECL unique is that they started by rethinking power, not just racks and real estate. FlexGrid gives them a programmable energy layer for AI that can blend any energy source on a site‑by‑site basis. This flexibility is what is needed to scale AI rapidly and sustainably.”
As AI‑related power demand accelerates worldwide, industry analysts project that incremental AI‑optimized servers will drive steep increases in data center electricity consumption, turning energy availability and mix into a strategic constraint for AI deployments. By decoupling power architecture from any single fuel and enabling location‑agnostic deployments, FlexGrid is designed to help customers stay ahead of these constraints while supporting their own sustainability and cost objectives over time.
“Hydrogen was an important starting point for ECL, but it was never the end state,” added Bachar. “We built our patents and architecture around the idea that power should be flexible. You should be able to plug in hydrogen where it’s abundant, natural gas where it’s ubiquitous, renewables where they are competitive, and still deliver the same high‑quality power to the data hall. FlexGrid is how we take that vision to market and put ECL at the center of the AI inferencing tornado.”
