01 Solution · Private cloud

Private cloud services in your own data centre, operated around the clock

Self-service, elasticity and automation are properties of the platform, not of the hyperscaler. We build your private cloud in your own data centre, in colocation or in German data centres, migrate your systems onto it and run it 24/7. Data, location and access stay with you.

Book a call Or read the whitepaper first

Platform: Proxmox, KVM or Kubernetes on Debian. All open source, no proprietary layer.
Location: Your own data centre, colocation, or German data centres that are ISO 27001 certified.
Operations: 24/7 with on-call cover, monitoring and alerting, patch and escalation management.
Exit: Configuration lives in Git repositories that belong to you. No lock-in, not to us either.

02 Starting point

On premise is not a step backwards, and not a guarantee either

Three sentences from almost every first conversation on this subject.

“For this workload the numbers no longer add up.”

Egress charges, cross-zone traffic and premium support tiers were rarely in the business case that justified the move. They surface once the volume grows.

“In our own data centre this takes days again.”

The business side is used to minutes. Without self-service and automation you trade a dependency problem for an efficiency problem, whatever the cost sheet says.

“The server sits in our building, so we are in control.”

Control means traceable access, documented configuration, a reproducible restart and the ability to change operating partner. Location alone delivers none of it.

The reason for going to the public cloud was rarely the location, it was the operating model. Bring workloads back and lose that model, and you have cut the bill and slowed the organisation down.

03 The solution

Private cloud as a service, on infrastructure you control

The building blocks a modern cloud provider offers, on infrastructure you own or control inside a European legal space. Each block is bookable on its own.

Location analysis and workload assessment

We assess your largest workloads one by one: public cloud, on premise cloud or hybrid. The result is a target architecture with costs and risks, not a decision of principle.

A platform with self-service and automation

Provisioning, scaling and configuration run through infrastructure as code and self-service interfaces instead of tickets. That layer separates a private cloud from classic server operations.

Virtualisation and Kubernetes where it pays

Proxmox or KVM for classic workloads, Kubernetes to CNCF standard where it pays. No Kubernetes by default: several of our largest projects deliberately run without it.

Governance and security inside the architecture

Access control, network segmentation and audit readiness belong in the platform, not in an add-on fitted later. Who holds administrative access is documented, providers included.

Managed private cloud, 24/7 with on-call cover

The call at three in the morning reaches us. Monitoring and alerting, patch and escalation management and ongoing advice belong to it. Not a portal that leaves you on your own again.

Hybrid cloud solutions for load peaks

Elasticity without holding capacity permanently: peaks run through a connection to a European data centre. Hybrid is the normal case here, not a compromise.

04 How it works

Consulting, transition, operations: one contract, one team, one responsibility

No patchwork of consultancy, systems integrator and operator. Whoever runs it in the end plans differently.

Phase 01

Consulting

Location analysis, assessment per workload, the decision between on premise and hybrid. It ends with a target picture including costs and risks, and the first go or no-go before anything changes.

Phase 02

Transition

We build the platform including its automation layer, then migrate service by service with a fallback path. Your systems keep running, and there is no cut-over date on which everything tips.

Phase 03

Operations

Managed operations at the service level your teams know from the public cloud: monitoring, security, FinOps and automation, 24/7, with on-call cover.

In projects so far, two to six months passed between breaking ground and a productive target environment, depending on the estate. We build nothing we would not want to operate ourselves.

05 Proof

Four organisations whose platform runs where it belongs

Four different starting positions: consolidation onto one private cloud, the way out of the public cloud, load peaks, heightened data protection requirements.

konversionsKRAFT Digital marketing, around 80 employees, SaaS platform with terabit data volumes. Consolidated onto a highly available private cloud with a redundant database cluster. Proxmox · Percona · GitLab

Innotech The application ran in Azure, with performance problems and a cost curve that only pointed upwards. Two months from breaking ground to the new production environment on private cloud, then 24/7 with patch and escalation management. Proxmox · Debian · HAProxy · PostgreSQL

Aerosoft Their own dedicated infrastructure, but brutal load peaks at product launches and on Black Friday. Availability permanently above 99.9 per cent, the complete stack from network to database from one source. Proxmox · HAProxy · Percona · Redis

Pro54 Marketplace for corporate health management, around 20 employees. HR data in a financial context, operated under heightened data protection and security requirements. Stack on request

06 Read on and check

Free whitepaper · 7 pages

Your Own Cloud in Your Own Datacenter

By Andreas Hankel, CTO onehundred. Download in exchange for your e-mail address, no sales call.

Get the whitepaper

07 To be honest

Four objections you are having right now

“Is a cloud of our own not a step back into the classic data centre?”

Only if the platform layer is missing. Self-service, elastic scaling and API automation are properties of the platform above the hardware, not of a hyperscaler. We build that layer, so your teams carry on as before, on infrastructure whose location and access you decide.

“We do not have a data centre of our own.”

You do not need one. We run the same platform in German data centres that are ISO 27001 certified, or in colocation. As a hosted private cloud you take the infrastructure from us and buy no hardware. If you do have a data centre, we bring the stack to you.

“Then we depend on you instead of the hyperscaler.”

Only if we pulled you onto a platform of our own. We do not. The stack is open source and close to CNCF, your configuration lives in Git repositories that belong to you, and any competent provider can take operations over. The price for us: we have to earn it again every month.

“We do not want to bring everything back.”

You should not. For most estates hybrid is the permanent target architecture, not a transition. We assess every workload on its own, and where one belongs in the public cloud we say so.

08 Fasttrack analysis

Nils HornkeCEO, onehundred

Start with the one workload that is giving you the most trouble

Name the one workload where cost, control or compliance is hurting. You get an assessment of whether a private cloud is the right home for it, and if not, we say so.

Book an intro call

15 minutes · not a sales call · book directly in the calendar

09 Frequently asked questions

What else you may want to know

What does a private cloud cost to run?

That depends on the number of systems, the availability required and your stack. You get a reliable figure after the first conversation, including the operating effort that cost models for owned infrastructure regularly leave out.

Do we need a data centre of our own?

No. We select a German data centre and run the platform there, or we work in colocation. If you have your own, we run the same stack on your site.

Where are the servers?

In German data centres, and in your own on request. Your data does not leave them.

Is onehundred ISO 27001 certified?

onehundred works to ISO 27001 and BSI compliant processes, and the data centres in use are certified. Certification of the company itself is not claimed.

Which workloads should we bring back from the public cloud?

The ones whose load is predictable and whose data has to stay inside one legal space anyway. Workloads with fluctuating load or a deep tie to a managed service are better left where they are.

Do we have to move to Kubernetes for this?

No. We run classic virtualisation on Proxmox or KVM just as happily. Kubernetes is a tool, not a goal.

Can AI workloads with sensitive data run on this platform?

Yes. GPU workloads, data pipelines and training or inference workloads run where the data has to stay anyway. That shortens approvals, because compute and data sit in the same legal space.

What happens if we want to change operating partner?

You take your Git repositories and carry on. That is why the stack is chosen the way it is: open source, documented, with no proprietary layer only we can operate.

10 Insights · Private cloud in your own data centre