Blog
From strategy to execution: how organisations can move data and workloads back on‑prem
The third part of our digital sovereignty series, this article moves from principles to execution. It explores what going back on‑prem actually looks like in practice, breaking the challenge down into data, tools and infrastructure. It outlines what is easiest to move, where organisations are likely to encounter limits, and how a realistic hybrid approach can help regain control without compromising security, capability or resilience.
Jump to:
- Finding pragmatic midway solutions
- From theory to practice: going back on‑prem
- Data
- Tools
- Infrastructure
- What we offer at Conscia
Finding pragmatic midway solutions
As organisations begin to explore what a more balanced and sovereign technology strategy might look like, many are already taking initial steps by diversifying their technology portfolios. While EU‑based alternatives to established cybersecurity vendors do exist, they are not yet consistently on par with leading providers, and most organisations are understandably reluctant to compromise when it comes to security capability.
But one thing already happening is that US companies are adding on-prem capability to their offerings to facilitate customer needs. Cisco, for example, has a subset of its products named Sovereign Infrastructure, designed to reduce reliance on centralised cloud services and provide greater control over critical components.
From theory to practice: going back on‑prem
But say a company decided to go back on-prem, how would that work in reality? We can split this into three aspects: data, tools, and infrastructure.
Data
When organisations look to move away from the cloud, data is often the easiest place to start. Everyday files stored in services such as OneDrive, Dropbox or Google Drive are relatively portable, and platforms like Nextcloud can offer a similar experience with little disruption. Regulation is also encouraging greater flexibility, with the Data Act, NIS2 and the AI Act pushing organisations to think more carefully about where their data is hosted.
More sensitive data raises tougher questions. European examples, such as France moving its Health Data Hub to a SecNumCloud‑certified provider and Norway exploring exit strategies for critical functions, show that repatriation is possible but usually limited to the most critical datasets. In practice, the aim is not a full cloud exit, but a balanced hybrid approach where different types of data are kept in the right place.
Securing IaaS for Cegal
Tools
If files are the easy part, tools and software as a service are far more difficult. This is where organisations often realise that their data is deeply embedded in the applications they use every day. While a file can usually be moved with little effort, data inside services such as email, time tracking or CRM systems is tightly bound to the platform itself.
Most organisations also underestimate how many cloud tools they depend on, from HR and invoicing to analytics and design. In these cases, the data sits entirely within the vendor’s environment, and the hosting decision is theirs, not yours. Even with the Data Act improving portability rights, moving away from a tool usually means replacing it altogether. For many organisations, this is where ambitions of a cloud exit meet the practical limits of the software ecosystem they rely on.
Infrastructure
Infrastructure is often the most portable part of a cloud setup. While organisations buy compute and storage from hyperscalers, the underlying technologies can usually run elsewhere. Workloads running on Kubernetes can be dynamically moved between on-prem Kubernetes clusters and clusters running at a European provider, offering far more flexibility than software tools.
However, infrastructure moves still come with trade-offs. Cloud platforms bundle services like firewalls, monitoring and proxies, which must be replaced or rebuilt if workloads are brought home.
A hybrid approach is therefore the most realistic option, with sensitive or performance-critical systems running locally and the rest remaining in the cloud. Lock-in, skills and complexity all mean that infrastructure migration requires careful planning about what should move and what is better left where it is.
What we offer at Conscia
At Conscia, we deliver infrastructure designed to support serious and demanding data‑centre environments, with security built directly into the network. We provide infrastructure‑as‑a‑service and can act as a European‑based data centre, either by building the environment for our customers or by supporting them in building it themselves.
This includes servers and storage, architected according to best‑practice principles so that all components connect reliably and efficiently. The network fabric is designed to support high‑bandwidth, performance‑critical workloads, and can be complemented with our managed detection and response services to provide ongoing security alongside the infrastructure.
In addition, we offer technical expertise on European cloud providers like Scaleway, which aim to replicate the core capabilities of hyperscalers while emphasising European data sovereignty and regulatory compliance.
IaaS
Experts in Amazon Web Services We are a certified AWS Advanced Consulting Partner and experts in Amazon Web Services (AWS). Our consultants have been approved for the AWS DevOps Co…
About the author
Lars Erik Braatveit
CTO, Conscia Norway
Lars Erik Braatveit has over 20 years of experience securing businesses, their systems, and services, including delivering security testing to some of the largest organizations in Norway. He currently works as Cyber Security Lead at Conscia Norway.
Related
