What Is It?
Managed Kubernetes provides fully managed Kubernetes clusters with automatic control plane management, version upgrades, and security patches. The geo-redundant, regional control plane is hidden from users and accessed only via the Kubernetes API. Create public or private node pools with customizable compute resources (vCPU or Dedicated Core), storage options (HDD/SSD), and networking configurations. Supports standard kubectl workflows via downloadable kubeconfig files.
Quick Facts
| Aspect | Details |
|---|---|
| Type | Managed Kubernetes (K8s) service |
| Control Plane | Fully managed, geo-redundant, regional, high-availability |
| Node Pools | Public (internet-exposed) and Private (isolated LANs) |
| Maximum Node Pools | 500 per cluster |
| Compute Options | vCPU servers or Dedicated Core servers |
| Storage | HDD or SSD with customizable sizes |
| Autoscaling | Horizontal node pool scaling (min/max configuration) |
| Management | Kubernetes Manager in DCD, Cloud API, kubectl |
Supported Kubernetes Versions (Examples)
| Version | Release Date | Available Date | Product EOL |
|---|---|---|---|
| v1.33 | 23 Apr 2025 | 25 Aug 2025 | 28 Jul 2026 |
| v1.32 | 11 Dec 2024 | 5 May 2025 | 28 Mar 2026 |
New node pools cannot be created on EOL versions; upgrade required first.
What You Can Do
Create and Manage Clusters
Deploy Kubernetes clusters with full administrator access to the Kubernetes API. Create up to 500 node pools per cluster, scale horizontally, and configure maintenance windows (4-hour window, day and UTC time configurable).
Configure Node Pools
Create public node pools (internet-exposed with load balancing) or private node pools (isolated, reachable only via private LANs ). Configure autoscaling with min/max node counts, labels, annotations, and availability zones.
Customize Compute Resources
Choose vCPU or Dedicated Core server types for worker nodes. Select number of CPU cores, RAM size, storage type (HDD/SSD), and storage size per node pool.
Network Integration
Attach private LANs to node pools, reserve static IPs for services, and configure network policies. Preserve source IP addresses for Ingress traffic when needed.
Automatic Updates
Control plane receives security patches during configured maintenance windows. Node pools receive both K8s and OS updates
Monitor and Log
Access integrated monitoring and logging for traffic, resource utilization, and security events. View read-only managed resources in DCD inspector. DNS Customize CoreDNS configuration for DNS resolution requirements.
API and Automation
Download kubeconfig files (YAML/JSON) via UI, API, or CLI for standard kubectl workflows. Use IONOS Cloud API (OpenAPI spec), Terraform provider, or Ansible for cluster and node pool management.
Best For
| Scenario | Why It Fits |
|---|---|
| Microservices architectures | Automated orchestration and scaling for containerized applications |
| CI/CD pipelines | Integrate with deployment workflows for continuous delivery |
| Multi-tenant applications | Namespace isolation and resource quotas for tenant separation |
| Hybrid cloud deployments | Private node pools connect to on-premises infrastructure via LANs |
| Auto-scaling workloads | Horizontal pod and node pool autoscaling handle variable demand |
| Stateful applications | Persistent volume support with block storage integration |
Consider Alternatives If
| If You Need... | Consider | Why |
|---|---|---|
| Enterprise Kubernetes with additional tooling | Red Hat OpenShift | Integrated CI/CD, monitoring, and security scanning |
| Simple container hosting without orchestration | Compute Engine with Docker | Lower complexity for single-container workloads |
| Private container image storage | Private Container Registry | Secure registry for Docker images with access control |
Key Considerations
Billing & Costs
- Control plane: Managed by IONOS (included in service)
- Node pools: Billed per worker node (CPU, RAM, storage)
- Server types: vCPU (cost-effective) or Dedicated Core (guaranteed performance)
- Storage costs: HDD or SSD volumes per node
- Additional costs: Load balancers, public IPs, data transfer, persistant Volumes
Limitations
- Maximum node pools: 500 per cluster
- Naming constraints: 63 characters maximum, alphanumeric start/end, may contain hyphens, underscores, dots, no spaces
- EOL versions: Cannot create new node pools on end-of-life Kubernetes versions
- Provisioning access: No access to target data center while nodes are provisioning or updating
- Maintenance window: 4-hour window for control plane and nodepool updates
- Control plane access: Hidden from users, accessed only via Kubernetes API
Management Options
- Kubernetes Manager (DCD): Visual interface for cluster creation, node pool management, kubeconfig downloads
- kubectl: Standard Kubernetes CLI with downloaded kubeconfig
- Cloud API: Programmatic access via REST endpoints (
/k8s/{k8sClusterId}/kubeconfig) - Terraform provider: Infrastructure-as-code for reproducible deployments
- Ansible: Automation for cluster and node pool configuration