Benchmarks
Managed Kubernetes benchmarks: provisioning time, scaling and features
Managed Kubernetes is assessed on what operators feel: how long a cluster takes to become ready, how fast nodes join under autoscaling, and whether load balancers and persistent volumes are integrated. Runs use the provider’s Terraform provider or CLI, so creation times include the full control-plane provisioning.
Updated:
Direct answer
Compare managed Kubernetes by control-plane creation time, node join time under autoscaling, conformance, and native integrations for LoadBalancer services and CSI storage. Hetzner does not offer managed Kubernetes; the other providers tracked here do.
What is measured
- Cluster creation time to Ready (minutes)
- Node join time during scale-out (seconds)
- CNCF conformance
- LoadBalancer, storage class and autoscaler integration
Tools
- kube-burner — Control-plane and scale testing
- Sonobuoy — Conformance testing
Provider capability matrix
Every value is based on provider documentation. Undocumented features are shown as unverified.
| Criterion | PortvMind | AWS | Azure | Google Cloud | DigitalOcean | Vultr | Hetzner |
|---|---|---|---|---|---|---|---|
| Kubernetes | |||||||
| Managed Kubernetes Provider-operated Kubernetes control plane. | Yes | Yes | Yes | Yes | Yes | Yes | No CCM and CSI driver for self-managed clusters only. |
| Kubernetes node autoscaling Worker node count scales automatically within configured limits. | Yes | Yes | Yes | Yes | Yes | Yes | Partial Through the upstream cluster autoscaler on self-managed clusters. |
| Kubernetes LoadBalancer integration Services of type LoadBalancer provision a provider load balancer automatically. | Yes Public and internal load balancers. | Yes | Yes | Yes | Yes | Yes | Partial Via the Hetzner cloud-controller-manager on self-managed clusters. |
| Kubernetes storage classes (CSI) Persistent volumes provisioned dynamically through a provider storage class. | Yes | Yes | Yes | Yes | Yes | Yes | Partial Via the Hetzner CSI driver on self-managed clusters. |
Metric definitions
| Metric | Unit | Better | Definition |
|---|---|---|---|
| Cluster creation | min | Lower | API create call to all nodes Ready. |
| Node join | s | Lower | Pending pod to new node Ready under cluster autoscaler. |
Verified results
Related research
Provider discovery
OpenStack cloud providers: what an OpenStack public cloud offers and how to evaluate one
What OpenStack-based public clouds provide, how their APIs and tooling compare with proprietary clouds, and a checklist for evaluating an OpenStack provider.