The European race to attract hyperscale data centers is being decided by factors that go far beyond electricity prices. Land availability, actual access to grid capacity, and delivery speed have become the three bottlenecks that separate project announcements from those that truly reach production. In this context, the partnership between Iberdrola and Echelon Data Centres — presented by Echelon itself as a new model for “hyperscale” development — aims precisely to address these three fronts in a coordinated manner.
The chosen approach is not a conventional commercial agreement. Both companies have established a joint venture, Echelon Iberdrola Digital Infra, with a participation structure where Echelon holds an 80% stake and Iberdrola the remaining 20%, through its digital infrastructure subsidiary CPD4Green. The declared goal: to build and operate large-scale data centers in Spain with a focus that combines site development and electrical connection from the outset, along with campus execution and subsequent operation.
A new “starting point”: the network first, then the building
Echelon’s message on LinkedIn is revealing: this model intends to solve “the three major constraints” of the sector —land availability, access to grid capacity, and delivery speed— and sets a standard for sustainable and scalable infrastructure. Behind this idea lies a reality known to any engineering or expansion team: data centers are no longer just blocked by urban licenses or CAPEX, but by electrical capacity and connection timelines.
Practically, the advantage of an alliance between an electric utility and a specialized developer is reducing the riskiest point: securing land with viable connection and guaranteed 24/7 energy. According to publicly available details of the agreement, Iberdrola provides exactly that: strategically located land with network connectivity and a commitment to supply clean electricity. Echelon, on its part, handles the most execution-intensive aspects: permits, design, marketing, and daily operations.
This division of responsibilities aligns with market trends: hyperscalers demand rapid capacity but also traceability over energy, schedules, and resilience. This requires “industrializing” the deployment process, avoiding critical infrastructure dependencies on late-stage integrations.
Madrid South: the first major campus and a marker of the size of the challenge
The first project announced under this platform is Madrid South, a complex covering 160,000 m² with a planned capacity of 144 MW of processing power and a guaranteed electrical connection of 230 MW. The timeline sets its operational start before 2030, with an estimated annual demand of around 1 TWh (1,000 GWh). Part of the supply would come from a solar PV plant on site, complemented by additional renewable energy from Iberdrola.
Beyond the headlines, these figures help contextualize the technical discussion:
- 144 MW IT is not a “big” data center; it’s a campus designed for massive loads and modular expansion.
- A connection of 230 MW demonstrates that the design incorporates margin for redundancy, peaks, and growth, including auxiliary infrastructure.
- An annual consumption of 1 TWh positions the project at the scale of a significant industrial user, with direct implications for the grid, energy planning, and supply agreements.
This underscores a key point: the bottleneck is no longer just in the building (which can be constructed with industrialized methods and specialized partners), but in the grid capacity and its availability over time.
Europe’s “largest alliance”: size, investment, and a signal to hyperscalers
Iberdrola has described this joint venture as the largest strategic alliance in Europe between an energy company and a data center developer. The planned investment exceeds 2 billion euros, focusing on creating a network of large-scale campuses in Spain. Meanwhile, economic sources suggest the plan could expand with additional projects and larger investments as connections and sites are consolidated.
The core message is clear: Spain aims to compete not just on cost but on capacity to deliver. For hyperscalers, the promise of “land + network + energy + operation” within a unified framework reduces friction, facilitates planning, and streamlines everything from site selection to go-live.
Technical implications: from “data center” to “network infrastructure campus”
This kind of partnership is also driving a conceptual shift: the data center is no longer just a building with servers but is understood as critical infrastructure linked to the electrical system. This requires making decisions from the outset that previously would have been resolved in later phases:
- Energy design and redundancy
24/7 availability and power stability are not guaranteed solely by UPS systems and generators. For campuses over 100+ MW, strategies include supply diversification, multiple connection points, and coordinated planning with the grid operator. - Renewable strategy and traceability
It’s not enough to “buy green” — major clients demand guarantees, PPAs, certificates, and often a sustainability narrative aligned with regulation and reporting requirements. - Modular delivery speed
Campuses are conceived as replicable units: phases, server rooms, electrical modules, scalable cooling. The goal is to activate capacity in stages rather than waiting for the entire project to be completed. - Permitting and schedule risks
Permits, environmental impact, electrical infrastructure, water/cooling, logistics, and connectivity all need early coordination to avoid surprises and rework.
Madrid as a market thermometer: “momentum” and capacity pressure
Echelon reports making “great progress” with projects in Madrid. The capital has become a hotspot for the industry due to its connectivity, business ecosystem, and digital demand, but also faces increasing pressure on the network and connection timelines.
This joint venture sends a signal: growth will no longer rely solely on land and financing, but on vertical integration between energy and campus development. If scaled as planned, this approach could influence future site negotiations, PPA structuring, and how a country’s competitiveness for hosting AI and cloud workloads is assessed.
Frequently Asked Questions
What does it mean for a data center to be “hyperscale,” and why is it measured in MW?
A hyperscale data center is designed for massive, modular growth, typically supporting large cloud platforms. MW represent the available electrical capacity for IT and auxiliary systems; in these projects, power is the best indicator of the true size.
Why is grid access now the biggest bottleneck for data centers in Spain?
Because demand for power is outpacing available connection capacity at certain nodes. Without a viable connection point and a clear schedule, projects can be delayed by years even if the building itself is “easy” to construct.
What does a partnership between an electric utility and a data center developer bring to a hyperscaler?
It reduces risk and timelines: land with connection, guaranteed 24/7 power, energy planning, and an operator to execute and manage the campus. It accelerates the “time-to-power” and “time-to-market.”
What are the implications of a 144 MW campus like Madrid South on consumption and sustainability?
It entails consumption on the order of hundreds or thousands of GWh/year, demanding renewable strategies, energy traceability, long-term PPAs, and efficiency designs (cooling, density, heat recovery where feasible).
via: Echelon Data Centers