Digital Sovereignty Needs Europe’s Next-Wave Data Center Markets

Digital sovereignty cannot be delivered through software, procurement rules, or cloud contracts alone. Europe also needs enough data center capacity, under the right legal and operational conditions, to host sovereign workloads. With capacity in the traditional data center hubs tightening, next-wave markets such as Hamburg, Vienna, and Barcelona are becoming essential to Europe’s digital sovereignty strategy.

Why digital sovereignty has become a capacity question

Europe’s sovereignty debate has moved from theory to practice. The EU’s Data Act aims to reduce switching barriers between cloud providers, while the proposed Cloud and AI Development Act would introduce a common sovereignty framework and aims to at least triple EU data center capacity within five to seven years. These policies matter because sovereign workloads still need physical servers, power, cooling, connectivity, and data centers in which to run.

Jurisdiction also matters. A server located in the EU does not by itself remove exposure to foreign law. The US CLOUD Act can require providers subject to US jurisdiction to produce data within their possession, custody, or control, regardless of where that data is stored. For sovereign workloads, organizations therefore need to assess both the physical location of the infrastructure and the legal and operational control exercised by the provider.

That distinction is important for GDPR as well. GDPR does not make Hamburg inherently “more compliant” than Amsterdam, or another EU/EEA location. Regional capacity instead gives customers more options to keep processing within the EU/EEA, choose providers whose legal structure fits their sovereignty requirements, and build geographic diversity.

FLAP-D remains critical, but it cannot carry all future growth

Frankfurt, London, Amsterdam, Paris, and Dublin remain Europe’s most important data center hubs. The problem is not their relevance; it is their ability to absorb the next wave of demand.

CBRE’s January 2026 outlook forecast European data center vacancy across primary and secondary markets to fall to 6.5% by year-end. Its midyear review tightened that forecast to 6%. Dublin illustrates the pressure: Knight Frank reported only 1.3% vacancy, with just 4 MW of available colocation capacity and almost 95% of capacity under construction already pre-let.

Power is increasingly the limiting factor. Grid constraints delay projects, while scarce powered land, higher costs, and more complex permitting make expansion harder in mature hubs. In some established markets, obtaining additional grid capacity can take years. While these challenges are not unique to established hubs, next-wave markets can provide additional routes to capacity where local conditions allow.

The underlying contradiction remains: Europe is asking for more workloads to be hosted under European control at the same time as suitable capacity in its traditional hubs is becoming harder to build. Digital sovereignty therefore has a supply-side problem. Procurement policy can define which providers and jurisdictions are acceptable, but it cannot create megawatts, land, grid connections, or new data halls. If compliant capacity is unavailable, stricter sovereignty requirements simply compete for an increasingly scarce pool of infrastructure.

Policy targets alone cannot solve this problem. Grid operators need to make power available, national and regional authorities need to provide workable planning conditions, and local authorities need to process permits within reasonable timeframes. Europe’s sovereignty ambitions therefore depend on coordination all the way from EU policy to individual infrastructure projects.

Next-wave markets are part of the digital sovereignty solution

This is why next-wave markets matter. Cities such as Hamburg, Stuttgart, Vienna, Lisbon, and Barcelona can provide additional routes to capacity beyond the traditional hubs, depending on local power, land and permitting conditions.

They can also offer practical advantages beyond capacity. Regional infrastructure can reduce latency for workloads that benefit from being closer to users or data sources, including some AI inference applications. For colocation customers, regional sites can also make physical access to hardware easier for installation, maintenance, or incident response.

The shift is already visible. The European Data Centre Association forecasts selected Tier-2 markets to grow at a 20.4% compound annual rate between 2024 and 2031, compared with 12% for Tier-1 FLAP-D markets. Madrid alone is forecast at 24.2%.

Colocation is especially relevant to this expansion because it separates facility operation from IT control. The operator is responsible for the building, power, cooling, physical security, and connectivity; the customer brings the servers, software, and data. Compared with building an on-premise facility, colocation removes much of the real estate and facility burden. Compared with using public cloud, it gives much more control to the customer.

For organizations and service providers with sovereignty requirements, regional colocation is a practical way to add capacity without having to build every site themselves. Colocation allows them to extend that footprint across several European markets while retaining control over their IT environment and data.

For digital sovereignty, a broader footprint also improves resilience by reducing dependence on individual markets and giving organizations more options to distribute workloads across independent locations.

The caveat is that a data center in a next-wave market does not automatically mean sovereign. Customers still need to examine the operator’s jurisdiction, ownership and control, access rights, subcontractors, and contractual model. Physical location is only the basis for sovereignty.

Sovereign workloads need somewhere to run

The scale of the challenge becomes clearer when growth rates are compared with absolute capacity. In the same EUDCA study, Tier-2 European markets supplied around 850 MW in 2024, compared with roughly 3,610 MW across Tier-1 FLAP-D markets. Next-wave markets are growing faster, but they still begin from a much smaller base.

That gap matters because hosting sovereign workloads still requires capacity at scale. Providers need enough data center capacity not only for primary workloads, but also for redundancy, growth, and geographic diversity. A policy that encourages more sensitive workloads to remain under European control will therefore increase demand for suitable European data center capacity.

The practical consequence is that digital sovereignty policy and physical capacity planning cannot be treated as separate agendas. If Europe wants more workloads hosted under European jurisdiction and operational control, it also needs to make sure that new data centers can actually be built.

Next-wave markets offer that opportunity, but only if the infrastructure there keeps pace. Grid reinforcement, predictable permitting, and access to suitable sites determine whether additional European-controlled capacity can be delivered at scale.

Digital sovereignty must be matched by physical capacity planning

Europe can define stricter sovereignty goals – but has to consider the infrastructure side. Traditional hubs will remain indispensable, but focusing on them for the next generation of sovereign cloud, AI, and digital public services is probably going to be difficult.

Digital sovereignty ultimately depends on choice: choice of provider, jurisdiction, location, and architecture. Organizations that want to implement this digital sovereignty need sufficient capacity. The next phase of digital sovereignty will therefore be determined as much by grid connections, permitting decisions, and investment in next-wave markets as by the rules written in Brussels.

As Managing Director of the AtlasEdge DACH Region, Georg Raiser is responsible for all aspects of the business in Germany, Austria, and Switzerland. He brings 20 years of professional financial experience across various industries. Before joining AtlasEdge, he held several roles in consulting and senior positions in the metal cutting sector. In 2018, he became CFO at Datacenter One, which was acquired by AtlasEdge in 2023. He holds a degree in business management from the University of Tübingen.