Which cloud for which purpose. EU data residency, vendor lock-in, cost — and why the question is rarely a technical one.
Category · DevOps & Infra
Three providers, one principle.
AWS, Azure and Google Cloud offer compute, storage and services on a rental model — billed by the minute, available worldwide. They differ in the detail, but at heart they do the same thing.
AWS is the broadest, Azure is strong in the Microsoft world, GCP leads on data and AI. Which one fits depends less often on the technology than on the context.
Why the question is rarely technical.
For German clients it's usually data residency that decides: are the data held in an EU region, who has access, what does the GDPR demand. Those answers come before the architecture, not after.
We pick the provider on data protection, the existing landscape and lock-in risk — and, where possible, keep a portable base of containers and IaC so we're not chained to one provider.
What gets underestimated.
Cloud isn't automatically cheap. Without cost control, egress fees, idle resources and managed-service markups quickly run out of hand.
And the lock-in is real: the deeper you use proprietary services, the more expensive the switch becomes. That's a deliberate trade-off, not an accident.
