Features
centron Kubernetes is a managed Kubernetes service: you provide your workloads, and centron operates the control plane and node pools. The following features are available in every cluster.
Key features
High Availability Control Plane
Operating self-hosted clusters requires specialist knowledge and time. centron takes care of managing the central services and settings for you. Optionally, you can replicate the central control plane components, thereby enabling an integrated failover mechanism.
You can find further details under High Availability.
Auto-scaling
Scale vertically and horizontally – using Cluster Autoscaling (CA) at the node pool level and Horizontal Pod Autoscaling (HPA) at the workload level.
See Enable autoscaling.
Complete Kubernetes API
Each cluster includes a control plane operated by centron and an API endpoint. This allows you to use any tool from the Kubernetes ecosystem – from kubectl and Helm to GitOps tools.
Automatic Upgrades & Patching
Minor patches and critical security updates are installed automatically within your upgrade window – on a rolling basis, node by node, and without any downtime.
See Cluster-Upgrade.
Integrated Monitoring & Alerts
The centron Metrics Agent is pre-installed on all worker nodes. You can monitor CPU, memory and bandwidth utilisation directly in Control Panel and set up alerts there too.
See View metrics.
Local NVMe storage
Each worker node comes with local NVMe storage – ranging from 25 GB to 400 GB, depending on the instance type. centron and S3 Object Storage are available for data that needs to be retained permanently.
Persistent volumes (CSI/PVC) are supported; use them to run stateful components directly in the cluster. Managed load balancers via LoadBalancer-type services are currently not available – set up external access via NodePort.
See Storage Features.
Further features
| Feature | Description |
|---|---|
| Free Control Plane | The Control Plane is included in the price. You pay only for your worker nodes. |
| Data sovereignty | Operates exclusively at the Bamberg data centre (BBG1), Germany. |
| Multiple Node Pools | A cluster can contain multiple node pools with different instance types. |
| Node Pool Autoscaling | Can be enabled per node pool, with configurable minimum and maximum numbers of nodes. |
| Surge Upgrades | During an upgrade, additional nodes are temporarily provisioned (maximum 10). |
| Upgrade Window | Freely selectable weekday and start time for a four-hour maintenance window. |
| VPC Integration | All resources in a data centre reside in the same VPC and communicate via private IP addresses. |
| Tags | Clusters, node pools and nodes can be organised using tags. |
| Projects | Clusters are assigned to a project and can be moved between projects. |