Kubernetes PaaS
Your applications. Our platform.
Get the Kubernetes clusters your team needs. basis keeps the clusters and underlying infrastructure running; your team configures and operates everything inside them.
- 99.7% API server SLA (PaaS)
- Standard Kubernetes APIs
- 24/7/365 on call
Is this for you?
- 01
Your team knows Kubernetes and wants its own clusters, with full cluster-admin.
- 02
You do not want to own servers, virtualisation, storage or the platform beneath the clusters.
- 03
You want a fixed monthly price for your clusters.
Operating boundary
Who runs what.
Want basis.no to operate your application as well? See Managed Kubernetes
| Layer | Kubernetes PaaS | Managed Kubernetes |
|---|---|---|
| The application itselfwhat it does, who uses it, its business data | You | You |
| Deployment and day-to-day operationsreleases, configuration, changes | You | basis.no |
| Application monitoring and incidentsalerts, on-call response, recovery | You | basis.no |
| Cluster add-onsingress, certificates, operators, monitoring stack | You | basis.no |
| The Kubernetes clusteravailability, provisioning, version upgrades | basis.no | basis.no |
| Platform and data centreHarvester, Rancher, storage, network, hardware | basis.no | basis.no |
| The SLA covers | The clusters and the infrastructure under them | The cluster and your application, in one SLA |
Full control of Kubernetes, no infrastructure to own
Teams that know Kubernetes well usually want their own clusters: their own operators, their own ingress, their own way of working. What they rarely want is to buy servers, run virtualisation and storage, and keep a data centre footprint patched.
Kubernetes PaaS splits the work at the cluster boundary. basis.no delivers and runs the platform the clusters live on. Inside the clusters, you are the administrator and decide what runs, how it is configured and which tools you use.
- Full cluster-admin – install operators and CRDs, choose your ingress controller, service mesh, monitoring stack and GitOps tooling
- No hardware or virtualisation – servers, Harvester, storage and networking are provided and operated by us
- Standard, conformant Kubernetes – nothing proprietary between your workloads and the Kubernetes API
- Our own platform – clusters and backups run in basis.no's own data centres
What the platform provides
Your clusters run on an open-source stack developed in the open by SUSE and the cloud-native community.
- No lock-in – standard Kubernetes APIs mean your workloads can run on any conformant platform
- Your tooling works – Helm, Argo CD, Flux, Terraform and your existing CI/CD pipelines work as they do anywhere else
Getting started
Getting your first cluster is a sizing exercise, not an infrastructure project.
- Tell us what you needNumber of clusters (for example test and production), node sizes, CPU, memory and storage.
- We provisionWe create the clusters on the platform and set up access for your team through Rancher.
- You take overYou get cluster-admin and install your own tooling: ingress, GitOps, monitoring, operators.
- Run and growWe operate the platform underneath. When you need more nodes or a new cluster, we agree the change with you.
What is included
These services are part of the monthly agreement.
| Service | What it means for you |
|---|---|
| Platform upgrades | Harvester, Rancher and Longhorn kept on supported versions, following a planned maintenance routine |
| Kubernetes upgrades | Your clusters kept on supported Kubernetes versions; we agree each upgrade window with you so you can test your workloads first |
| OS and security patching | The operating systems under the platform are patched by us |
| Cluster provisioning | New clusters and nodes created on request, within your reserved capacity |
| Volume backup and restore | Persistent volumes backed up on schedule, mirrored to a secondary location, restored on request |
| 24/7/365 monitoring | Hardware and platform services monitored around the clock |
| 24/7/365 support | Platform engineers on call, also at night and at weekends |
| Capacity follow-up | We notify you before you reach your reserved storage |
| Service desk | One place to report incidents and request changes |
The 99.7% SLA covers the Kubernetes API server of your clusters. It does not cover downtime caused by components or configuration you install in the clusters, load balancing or VPN on your side, or planned maintenance windows.
Fixed price per month
You reserve the capacity you need and pay a fixed monthly price for it.
Platform serviceInfrastructure, cluster provisioning, patching, monitoring, backup and support, as described above.
Compute capacityCPU and memory reserved for your cluster nodes.
Storage packageDisk capacity and backup sized to your data volume.
Capacity can be re-allocated between your clusters, or increased, by agreement.
Common questions
What does full cluster-admin mean in practice?
You can create namespaces, define RBAC, install operators and CRDs, and choose your own ingress controller, service mesh, monitoring and GitOps tools. basis.no does not decide what runs inside your clusters.
How is this different from Managed Kubernetes?
On Managed Kubernetes, basis.no installs a cluster in your tenant, deploys your application and operates both under one SLA. On Kubernetes PaaS, you operate everything inside your clusters, and basis.no keeps the clusters and the platform underneath running.
Who upgrades Kubernetes on our clusters?
basis.no does. Kubernetes releases a new minor version roughly three times a year, and each is only supported for around 14 months. We plan each upgrade with you and carry it out in an agreed window, so you have time to check that your operators and workloads are compatible first.
Are the clusters dedicated to us?
Yes. Your clusters are yours; no other customer's workloads run in them. They run on basis.no's shared Harvester platform.
Do you back up our databases?
We back up the persistent volumes. Application-level backups, such as consistent database dumps, are your responsibility, because only you know how your application must be restored.
Tell us what you want to run.
Tell us how many clusters you need and roughly how large. We come back with a proposed setup and a fixed monthly price.