Oasis Stack Oasis Stack A CloudHost Initiative
EN
Talk to us
Migrating from Red Hat

Keep the containers.
Lose the subscription tier.

Oasis Stack replaces OpenShift and the cloud management platform above it with one platform that delivers Kubernetes, the catalogue, the governance and the commercial layer — and takes the RHEL subscription footprint that tier carried with it.

Talk to an architect How it is priced
Your images stayKubernetes nativeFlat fee, banded by scale
Containers are portable
Your images and manifests are standard. What changes is the control plane they land on.
Kubernetes without the suite
Managed clusters from the catalogue, governed like every other service.
Infrastructure entitlements drop
The RHEL subscriptions that platform tier carried come off. Guest licensing is unchanged.
One platform, not a stack
Catalogue, tenancy, metering, billing and governance in one product.
The situation

You bought Kubernetes and received a subscription estate.

OpenShift solves a real problem, and it arrives with a platform tier, an operator ecosystem and an entitlement model that grows with every node you add. For an organisation running a handful of clusters that is a fair trade. For one trying to run a cloud — with tenants, a catalogue, chargeback and a commercial layer — it is a platform underneath the platform you actually needed, and both of them are licensed.

What moves and what stays

Standards stay. The tier above them goes.

Container workloads are portable by design, which is what makes this migration tractable.

Container images
Stay
Standard OCI images run unchanged. Nothing is rebuilt to a proprietary format.
Kubernetes manifests
Stay
Standard Kubernetes resources carry across. Operator-specific and Route-style objects are the ones to review.
Your CI/CD
Stays
Pipelines target a cluster endpoint. The endpoint changes; the pipeline usually does not.
Team knowledge
Stays
Kubernetes is Kubernetes. What your platform team knows continues to apply.
OpenShift control plane
Replaced
Managed Kubernetes is delivered from the Cloud Foundation catalogue, provisioned and governed like every other service.
The cloud management platform
Replaced
Tenancy, self-service, catalogue, metering, billing and governance come from Oasis Core rather than a separate management tier.
RHEL at the infrastructure layer
Replaced
The entitlements the platform tier required — control plane, infrastructure and worker nodes carrying OpenShift — come off with it.
RHEL inside customer VMs
Stays
A guest operating system is licensed by whoever runs it. If your customers run RHEL in their virtual machines, those subscriptions continue exactly as they are. This changes the infrastructure layer, not the guest.
The commercial model
Replaced
Per-node subscription becomes a flat fee per deployment, banded by the scale of the estate. Adding clusters and nodes inside a band does not re-price it. How a band works. Model your current licence bill.
What arrives with it

The layer OpenShift was never trying to be.

This is not a cheaper Kubernetes distribution. It is the cloud business around it.

Kubernetes as a catalogue item
Clusters ordered by a team, provisioned with policy already applied, metered and charged back.
More than containers
Compute, storage, network, databases, GPU and backup in the same catalogue, governed identically.
Commercial layer included
Quoting, contracts, subscriptions, metering, invoicing and partner margins — if you are selling, not just running.
Governance across everything
One policy and audit model reaching containers, virtual machines and marketplace services alike.
AI on the same platform
GPU pools and AI services governed and billed like every other service, not a separate stack.
Amin across the estate
Cluster health, capacity and cost read alongside everything else you run.
How the migration runs

Run both. Move namespace by namespace.

Container workloads make this easier than a virtualisation migration: the two platforms can run side by side, and applications move a namespace at a time with a pipeline change. The OpenShift estate shrinks as the new one grows, and the subscription comes off when the last workload has left.

Deployment architectures
1
Assess
Which workloads use OpenShift-specific constructs — Routes, Operators, SCCs, BuildConfigs — and what each becomes.
2
Stand up
Managed Kubernetes from the catalogue, with policy, quotas and chargeback applied from day one.
3
Move
Namespace by namespace, pipeline by pipeline, with both platforms serving during the transition.
4
Retire
The OpenShift estate and its entitlements come out when the last namespace has moved.
What this does not do

Where the work actually is.

Container portability is real, but it is not the same as zero effort — and the parts that take effort are predictable.

OpenShift-specific objects need work
Routes, SecurityContextConstraints, BuildConfigs and Operator-managed resources have Kubernetes equivalents, but they are a conversion, not a copy.
Support relationships change
Moving off a Red Hat subscription means moving the support path with it. That is a commercial decision as much as a technical one.
It does not re-architect applications
A workload that struggles on OpenShift will struggle on any Kubernetes. Modernisation is a separate programme.
Guest licensing does not change
The saving is at the infrastructure layer. RHEL running inside a customer’s virtual machine is still that customer’s subscription, and nothing here alters it — a CFO modelling this should count the platform tier only.
The other one
Running both? Migrating from VMware covers what is replaced there, what stays, and where the work actually is.
Migrating from VMware
Open. Modular. Intelligent. Limitless.

Bring us your cluster inventory.

We will map what OpenShift is doing for you today against what the platform replaces, and what your entitlement footprint becomes.

Talk to an architect How it is priced