Managed Kubernetes: Production Defaults Built In
Deploy a production Kubernetes cluster with networking, ingress, Helm, and storage pre-configured. HPA scales pods, the platform adds worker nodes as capacity is reached, and you pay only for the resources your cluster uses.
CNI, ingress, Helm, CoreDNS, storage provisioners, and monitoring are ready when the cluster starts.
Namespaces isolate workloads. RBAC service accounts scope access across multi-service deployments.
HPA adds pod replicas on CPU and memory thresholds. The platform adds worker nodes as capacity fills.
Run HA control planes behind API load balancers when you need production-grade resilience.
When to Use Managed Kubernetes
Pick the option that fits the workload. The recommended route covers most production cases; the two alternatives exist for specific scenarios.
Recommended · HA Kubernetes
Multi-master with HA
You need real Kubernetes APIs (Helm, CRDs, ingress, HPA) and the cluster has to survive node maintenance and rolling upgrades.
See HA topologyAlternative · Single control plane
Single control plane
Development, staging, or internal workloads that need Kubernetes APIs but not multi-master resilience.
See single-control-plane presetAlternative · Cloud VPS
Roll your own cluster
You need to assemble a custom Kubernetes distribution or tune the control plane outside the managed preset.
Compare Cloud VPSCluster Architecture, Pre-Configured
Select the preset, then provision the control plane, workers, ingress, persistent storage, and Helm tooling from the dashboard. See the provisioning flow in the walkthrough above.
- CoreDNS and ingress come ready for routing.
- Persistent storage provisions with the cluster.
- Metrics Server supports HPA from day one.
Included from first deploy
- Weave CNI: pod-to-pod networking across every node.
- Ingress controllers: HAProxy, Nginx, or Traefik with SSL.
- NFS dynamic storage: automatic volume provisioning.
- Helm preinstalled: chart-based releases from deploy one.
- K8s Dashboard & K9s: visual and CLI management.
- GlusterFS (optional): distributed storage for replicated volumes.
Choosing the Right Topology
Select a preset at install. Both paths include the same networking, ingress, storage, and dashboard tooling. The difference is control-plane resilience and resource footprint.
Topology is a cost and resilience decision.
Start with a single control plane for development or staging. Choose HA when the API endpoint, etcd quorum, and upgrade path need to survive node maintenance.
Price the topology before you launch. Model either preset with the calculator, then set scaling limits with Autoscaling & Cost Controls.
Pay for What Your Pods Consume
Cloudlet billing follows the limits you set.
Usage-based billing. You pay for the RAM and CPU your pods consume. The platform allocates resources within configured limits, and billing follows actual usage.
Burst without overpaying. Set generous scaling limits for traffic spikes. Idle periods bill only for storage and network.
Real-time visibility. Per-node resource and cost monitoring. Dashboard usage equals invoice usage. See Observability.
Build and cap the cluster
Cloudlet limits set on this screen cap the bill. Pods burst within them, and idle periods accrue no charges.
How a Traffic Spike Scales Your Cluster
A traffic spike triggers a sequence through both layers. Kubernetes reacts first, then the platform supplies capacity.
-
Request load rises
Pod CPU or memory crosses the HPA threshold you set on the deployment.
-
HPA adds pod replicas
Kubernetes schedules new replicas. If existing nodes have headroom, they fit. If not, pods stay
Pendinguntil capacity arrives. -
Cloudlets expand the existing node
The platform vertically scales the worker container’s RAM and CPU within the per-node ceiling you set. No restart, no eviction. VPA can also adjust per-pod resource requests on the same metrics stream.
-
A new worker node joins
When per-node ceilings are reached, the platform adds a worker node behind the cluster’s load balancer;
Pendingpods schedule onto it within seconds.
You pay only for the cloudlets actually consumed.
Cluster Protection and Zero-Downtime Upgrades
Feature availability matrix for Kubernetes backup, recovery, and upgrade capabilities.
Velero scheduled backup
Back up Kubernetes resources and persistent volumes on a schedule or timed interval.
Cluster-to-cluster restore
Restore to the same cluster or migrate state to a different one for disaster recovery or blue-green upgrades.
Zero-downtime component upgrades
Upgrade via the add-on system. Compatibility checks run before major version changes.
Common Kubernetes Questions
What is included with Togglebox managed Kubernetes hosting?
It includes preconfigured Kubernetes components (ingress, DNS, storage provisioning, Helm, dashboard), plus optional multi-master and add-ons for monitoring and tracing.
Which ingress controllers can I choose?
You can use HAProxy, Nginx, or Traefik ingress controllers based on your deployment needs.
Do you include Helm?
Helm handles one-click deployment of popular applications.
What DNS is included?
CoreDNS provides in-cluster DNS resolution. Pods and services resolve each other by name automatically, so you don't manage DNS entries manually within the cluster. External DNS for public endpoints is handled separately through your domain records.
What storage options are available for persistent volumes?
Persistent volume provisioning is supported. Optional storage choices include an NFS provisioner and a GlusterFS storage cluster.
What CNI/network plugin is used?
Weave CNI is used for internal networking.
Do you support multi-master Kubernetes clusters?
Multi-master with API load balancers ships as an optional configuration.
Do you automate Kubernetes installation and updates?
The platform automates Kubernetes installation, configuration, and updates, and provides additional cluster components.
How does scaling work (vertical scaling vs adding workers)?
Vertical scaling is automated based on load. Worker scaling can be automated or manual, depending on your setup. For baseline vs burst capacity, see Autoscaling & cost controls.
What monitoring and tracing options are available?
Metrics server is available, with optional Prometheus/Grafana for monitoring and Jaeger for tracing. See Observability for visibility patterns.
Should I use Kubernetes or App Hosting architecture presets?
Choose Kubernetes when you need orchestration primitives and a multi-service platform. Choose App Hosting presets when you want simpler PaaS clustering without Kubernetes overhead. Compare both in the pricing calculator.
Can I manage cluster spend with cloudlet limits?
Set baseline and max limits per node group, then tune worker scaling to stay within budget guardrails. Pair those limits with usage alerts so cost changes are visible early.
No matching questions found.
Test your stack on metered hosting
- Meters real usage: baseline plus actual use, not a fixed tier
- Turnkey clustering: traffic reroutes once clustering is enabled
- From Git push to live: in minutes, not hours
- Engineers on call: real support around the clock
Start a real trial environment
Deploy your app, test the ceiling, and estimate cost before you commit.