When Data Sovereignty Trumps the Cloud: RIPE's €5 Million Pivot

The Unraveling of a Cloud-First Bet

RIPE NCC, the regional internet registry responsible for IP address allocation across Europe, the Middle East, and parts of Central Asia, is unwinding a multi-year cloud migration that once seemed inevitable. Managing Director Hans Petter Holen announced the shift in a blog post and presentation at last month's General Meeting, citing geopolitical risk tied to US-based hyperscalers as the catalyst. The decision marks one of the more concrete examples of European infrastructure operators pulling back from American cloud platforms in the wake of heightened regulatory and political uncertainty.

At DailyTechWire, we've tracked similar reversals across financial services and public-sector IT in Frankfurt and Paris, but RIPE's case is notable for two reasons: the organization is a critical internet governance body, and it now faces the compounding problem of hardware that aged out while capital spending was diverted to cloud operating expenses. The result is a €5 million greenfield deployment slated for completion by 2028 - a figure that effectively resets capital expenditure to pre-2020 levels.

The Hardware Debt Comes Due

When RIPE adopted its cloud-first posture, the trade-off was explicit: shift from capital investment in on-premises infrastructure to operational spending on managed services. That model deferred hardware refresh cycles and assumed long-term cloud tenure. Now that the strategy is being abandoned, the organization is left with equipment that has reached or exceeded end-of-life, and no immediate replacement pipeline.

According to Holen, the overhaul extends beyond simple server replacement. RIPE must rethink its datacenter footprint - both the number of facilities and their geographic distribution - while minimizing interdependencies to allow future expansion. Geographically redundant storage and backup systems are also on the list, as are decisions around virtualization platforms that reduce vendor lock-in risk.

The scope of the rebuild reflects a broader reality for mid-tier infrastructure operators: reversing a cloud migration is not a return to the previous baseline. Stakeholder expectations around uptime, security, and resilience have risen, and the tooling and architectural patterns that defined pre-cloud deployments are no longer considered adequate. RIPE is effectively building a new on-premises stack from scratch, one that must meet the performance and availability benchmarks set by the hyperscalers it is leaving behind.

Geopolitical Risk as the Trigger

While Holen's announcement did not enumerate specific incidents, the timing aligns with a wave of European reassessments of US cloud dependency. The arrival of a new US administration in early 2025 amplified concerns over data sovereignty, extraterritorial jurisdiction, and the potential for unilateral policy shifts that could disrupt access to cloud services or data held by American providers.

For an organization like RIPE, which manages a foundational layer of internet infrastructure, the risk calculus is different from that of a typical enterprise. A disruption to IP address allocation or WHOIS database services would have cascading effects across the region. The decision to exit the cloud is less about cost optimization and more about ensuring continuity in a scenario where geopolitical friction could compromise access to third-party platforms.

This is not an isolated view. Over the past eighteen months, we've observed a pattern of European public-sector and critical-infrastructure operators favoring on-premises or hybrid architectures, often backed by domestic or EU-based cloud providers. RIPE's pivot is consistent with that trend, though it is among the few to make the shift public and to quantify the financial impact.

A Membership Divided on Who Pays

The infrastructure rebuild is only part of RIPE's challenge. The organization must also navigate a fractious membership debate over how to fund the €5 million investment. At the General Meeting, members voted on a proposal to replace the existing flat-fee structure with a sliding scale tied to the volume of internet number resources each member holds. Under the proposed model, 74 percent of members would have paid less, with larger holders absorbing a greater share of costs.

The vote failed - narrowly. Of 19,415 eligible members, only 3,049 cast ballots, a turnout of 15.7 percent. The result was 51.1 percent in favor of retaining the flat fee, a margin of just 35 votes. RIPE described the outcome as unexpected and devoted a follow-up blog post to dissecting possible causes.

One factor may have been communication friction. During the consultation process, RIPE floated multiple fee structures, some of which included base fees and additional charges that were later dropped after member feedback. Not all members appear to have tracked these revisions, leading to confusion about what the final proposal entailed. A persistent misconception was that higher fees would translate into greater voting rights - a notion RIPE explicitly rejected but which continued to circulate.

Another explanation is philosophical. Some members favor an egalitarian fee structure regardless of resource consumption, viewing differential pricing as inequitable. This mirrors debates in other member-funded technical organizations, where flat fees are defended as a form of institutional solidarity even when they shift costs away from the heaviest users.

The Equity Versus Equality Fault Line

The vote exposes a deeper tension within internet governance communities: whether contribution should be proportional to benefit, or uniform across participants. The sliding-scale model was framed as an equity measure, aligning fees with resource allocation and, implicitly, with the economic value members derive from RIPE services. The flat-fee model, by contrast, treats all members as equals regardless of their footprint.

This is not a new debate, but it has taken on sharper edges as infrastructure costs rise and the composition of membership diversifies. Smaller ISPs and newer entrants tend to favor progressive pricing, while established players with large allocations often prefer flat fees that cap their exposure. The fact that the vote was so close - and that turnout was still below 16 percent - suggests the issue is far from settled.

RIPE's post-vote reflection acknowledged that the organization may have misjudged the appetite for tiered pricing. The long-standing demand for a sliding scale, visible in prior consultations, did not translate into a voting majority. Whether this reflects satisfaction with the status quo or simply the inertia of low-engagement governance is unclear.

What the Pivot Signals for Regional Infrastructure

RIPE's cloud exit is a data point in a broader realignment of how critical digital infrastructure is provisioned in Europe. The calculus that drove cloud adoption - cost savings, elasticity, offloaded operations - has been complicated by regulatory fragmentation, geopolitical volatility, and the growing recognition that dependency on a handful of US platforms carries strategic risk.

For infrastructure operators, the lesson is that cloud-first strategies are not irreversible, but unwinding them is neither cheap nor simple. RIPE's €5 million rebuild is a mid-tier example; for larger registries or financial-market utilities, the cost and complexity would be higher. The decision also underscores the importance of maintaining in-house technical capacity. Organizations that fully outsourced infrastructure engineering during the cloud era may find themselves poorly positioned to manage a return to on-premises deployment.

At the same time, RIPE's experience highlights the governance challenges that accompany major infrastructure decisions. A narrow vote with low turnout does not provide a strong mandate for either fee structure, and the organization will likely face renewed pressure as the €5 million spending program unfolds. Whether future votes will draw higher participation - or whether RIPE will need to revisit the charging model - remains an open question.

The 2028 Deadline and What Comes Next

RIPE has set a 2028 target for completing the greenfield deployment. That timeline implies a multi-year procurement, build-out, and migration cycle, with parallel operation of legacy and new infrastructure during the transition. The organization will need to balance continuity of service with the urgency of replacing end-of-life hardware, a scheduling problem that becomes more acute as equipment ages.

The virtualization platform decision is particularly consequential. RIPE's stated goal is to limit vendor lock-in, which suggests a preference for open-source or standards-based hypervisors over proprietary stacks. That choice will shape both the technical architecture and the long-term cost profile, as licensing and support models vary widely.

Geographically redundant storage and backups add another layer of complexity. RIPE must select datacenter locations that are politically and operationally independent, minimizing the risk that a single jurisdiction or infrastructure failure could compromise multiple sites. In practice, this means balancing considerations of latency, cost, regulatory environment, and physical security - a multidimensional optimization problem that few organizations solve perfectly on the first pass.

The broader question is whether RIPE's pivot will prove to be an outlier or a leading indicator. If geopolitical risk continues to rise, other regional internet registries and critical infrastructure operators may follow a similar path. If the current wave of uncertainty recedes, RIPE's decision may look like an expensive hedge. Either way, the organization has committed to a course that will define its operational posture for the next decade, and the costs - financial and technical - are already locked in.