• News

Cloud Repatriation for strategic infrastructure governance

Reading time6 min

For almost a decade, the Public Cloud has been presented as a universal solution for flexibility, scalability and reduced operating costs. Thousands of companies have migrated critical workloads to public solutions, but operational reality has proven more complex. According to a Barclays survey conducted at the end of 2025, a growing number of companies worldwide are planning to move back: as many as 86% of the CIOs surveyed would like to move data and applications from the Public Cloud back to on-premises or private infrastructures

This phenomenon is known as Cloud Repatriation

What is Cloud Repatriation?

Cloud Repatriation is the deliberate migration process of data, applications and workloads from Public Cloud providers to locally installed infrastructures, private data centers or hybrid Cloud environments. 

However, it is important to note that planning a repatriation does not mean wanting to abandon the Public Cloud altogether; in most cases, it does not involve migrating the entire application portfolio. Today's approach, often referred to as Smart Cloud, is selective and strategic: it places each workload where it is most efficient, cost-effective and, above all, secure. 

This represents a significant paradigm shift compared with the narrative of recent years: it is no longer a question of choosing between Public Cloud and on-premises solutions, but rather of right-sizing** workloads,** making a considered decision that takes multiple factors into account for each individual application. 

Which Workloads Should Be Migrated On-Premises?

A well-considered Cloud Repatriation strategy involves clear segmentation of the different workloads, based on technical, economic and regulatory criteria. Some workloads are natural candidates for returning to private or local infrastructures, for example: 

  • Highly sensitive data and applications. Critical databases, identity management systems, intellectual property repositories or financial information benefit from direct oversight and granular access controls.
  • Low-latency and high-reliability applications. Industrial systems, financial trading platforms, real-time IoT systems and core ERP platforms cannot tolerate transmission delays. The physical proximity of the infrastructure reduces latency and ensures the required responsiveness.
  • Workloads with predictable and stable costs. Persistent databases, document repositories and applications with consistent consumption have much more predictable costs on a private infrastructure.
  • Highly customized systems. When a company has invested, for example, in dedicated Kubernetes clusters or highly specialized infrastructures, managing them in a private environment provides complete control over configuration, patches and updates.
  • Workloads with stringent audit requirements. Applications subject to regulations such as DORA, NIS2, GDPR or Swiss data protection laws require complete traceability and verifiable audit trails. A private infrastructure simplifies verification and compliance.

When to Consider Repatriation: Signals and Benefits

Not every company needs a total repatriation, but there are signals indicating when it is time to start considering it: 

  • Unexpected Cloud costs, with bills increasing significantly without proportional increases in the service provided;
  • Perceived latency in applications that are sensitive to transmission delays;
  • Stringent regulatory requirements (DORA, FINMA, NIS2) that make compliance difficult;
  • Critical dependence on a single provider, with the risk of vendor lock-in.

When these signals emerge, repatriation offers at least three concrete benefits, including: 

  • Predictability and cost reduction (TCO). For stable workloads with predictable consumption, a private Cloud infrastructure reduces costs and, above all, their variability.
  • Performance and latency reduction. Applications sensitive to delays, such as those in the financial sector, real-time industrial systems and IoT platforms, benefit from the physical proximity of a local infrastructure.
  • Vendor lock-in mitigation. By keeping critical workloads on private infrastructure, dependence on a single provider is reduced.

Security, AI, Control and Data Sovereignty

Cloud Repatriation profoundly transforms several aspects of a company's infrastructure ecosystem: its overall security posture, on the one hand, but also the adoption of new proprietary technologies based on Artificial Intelligence, given that the costs of major international players are rising and data security is certainly a critical issue. This is why, as Tinext Cloud, we offer AI Colocation services with 100% “Swissness”. Closely related to this is the issue of data sovereignty

In the Public Cloud, choosing a specific geographical region does not necessarily mean that all data, its copies and metadata remain within that same jurisdiction. The actual location depends on the provider, the service used and the configuration adopted and may involve, in addition to primary data, replicas, backups, logs and metadata that may be processed or stored in other geographical areas. 

A repatriation strategy guarantees certain data residency within specific geographical and jurisdictional boundaries. In 2025, Gartner coined the term Geopatriation precisely to describe this evolution: geopatriation is a direct response to geopolitical and regulatory factors, from international instability to the impact of laws outside the EU, such as the US Cloud Act. 

The Swiss Context: Regulations, Risks and Opportunities

For Swiss companies, Cloud Repatriation takes on particular strategic importance. The current regulatory framework requires extremely rigorous management of personal and sensitive data, with stricter traceability and breach notification obligations. In Switzerland, data protection authorities have recently expressed strong reservations about the use by public bodies of SaaS cloud services provided by large international operators for particularly sensitive data, sending a clear signal to the private sector about the importance of digital sovereignty. 

Indeed, according to a Swiss Federal Statistical Office survey, 58% of Swiss companies would cite dependence on hyperscalers as their main security concern, but most of them do not have structured exit plans in their contracts or the skills required to manage migrations. This creates a risk situation: awareness of dependence, but a lack of concrete strategies to mitigate it. 

This is why an increasing number of Swiss companies are adopting hybrid architectures, where critical and regulated workloads, such as those subject to FINMA requirements in the finance sector, reside on sovereign infrastructures within Switzerland, while less sensitive applications continue to leverage the global Public Cloud. This approach requires both technical and regulatory expertise, driving rigorous planning that makes it possible to balance efficiency, control and regulatory compliance.  

The Role of Modern Private Infrastructure

Modern private infrastructures, based on virtualization such as HPE VM Essential, offer agility and flexibility comparable to the Public Cloud, while providing benefits such as improved data governance, enhanced security and predictable costs. 

For Swiss companies, this represents a strategic opportunity: combining modern private infrastructure, strict governance and meticulous regulatory compliance creates a significant competitive advantage. Rather than being subject to dependence on global hyperscalers, it is possible to build an architecture that reflects the company's needs in terms of sovereignty, control and reliability.