Cloud Engine Containers
Kubernetes
Kubernetes doesn’t have to be complicated. With Bamboozle Cloud Engine Containers, we take care of the hard parts of cluster management, so you can focus on building and scaling your applications with total confidence.
- Control plane
- HA
- Node autoscaling
- Auto
- Persistent volumes
- NVMe
- DC regions
- 3
Key Capabilities
What You Get
Production-grade Kubernetes infrastructure managed by Bamboozle — from control plane to storage, networking, and observability.
-
HA Control Plane
Master nodes run in Active/Active high availability mode. The control plane is managed by Bamboozle — no single point of failure, no manual etcd backups.
-
Cluster Autoscaler
Worker groups scale up automatically when pods are stuck pending, and scale down when nodes sit idle. Set min/max per group — the cluster handles the rest.
-
Cilium CNI — eBPF Networking
Cilium replaces iptables with eBPF for packet routing and network policies. L3/L4/L7 policies, native load balancing, and Kubernetes NetworkPolicy support out of the box.
-
NVMe Persistent Volumes
Dynamic PVC provisioning backed by Bamboozle NVMe storage. Volumes exist independently of pods — data survives pod restarts, rescheduling, and node replacement.
-
LoadBalancer + Floating IP
Create Kubernetes LoadBalancer services and assign a Floating IP for stable public exposure. Reassign the IP instantly for failover without DNS changes.
-
Integrated Prometheus + Grafana
Enable integrated monitoring at cluster creation. Prometheus, Alertmanager, and Grafana deploy automatically — cluster health, node metrics, and pod utilisation ready in minutes.
-
Non-Disruptive Updates
Kubernetes version upgrades are rolling — worker nodes are updated one by one with data availability unaffected. Upgrade via the Cloud Control portal with one click.
-
kubeconfig Access
Download a kubeconfig file from Cloud Control to connect kubectl and any standard Kubernetes tooling — Helm, ArgoCD, k9s, Lens — directly to your cluster.
-
Multiple Worker Flavors
Mix General Compute, Compute Optimised, Memory Optimised, and High Frequency node groups within a single cluster. Assign pods to specific node groups with labels and affinity rules.
Worker Node Groups
Mix Flavors. Scale Independently.
Each worker group has its own VM flavor, autoscaling policy, and node count — so you can run CPU-intensive jobs alongside low-latency workloads in the same cluster.
-
Autoscaling
General Compute Pool
Web apps · APIs · CI/CD
-
Autoscaling
Compute Optimised Pool
CPU-heavy · HPC · build nodes
-
Fixed size
High Frequency Pool
Low latency · gaming · trading
Integrated Observability
Prometheus, Alertmanager & Grafana — Ready at Cluster Creation
Enable integrated monitoring when you create your cluster. The full Prometheus stack deploys automatically — no Helm charts to configure, no storage classes to provision manually.
- Prometheus scrapes all cluster and node metrics automatically
- Pre-built Grafana dashboards for cluster health, node utilisation, and pod metrics
- Alertmanager routes alerts to email, Slack, or your webhook of choice
- Export metrics to your own monitoring stack via the Prometheus remote write endpoint
- All data stored on NVMe persistent volumes — survives control plane restarts
Technology Stack
What Powers It
Standard upstream Kubernetes — all the tools you already use work out of the box.
- CNCF-Compatible Kubernetes
- Cilium CNI (eBPF)
- Dynamic PVC Provisioning
- Kubernetes NetworkPolicy
- L3 / L4 / L7 Policies
- LoadBalancer Services
- Floating IP Assignment
- Prometheus
- Alertmanager
- Grafana
- SSH to Nodes
- Rolling Version Upgrades
- Multi-worker-group Clusters
Getting Started
From Zero to Running Cluster
Clusters are ready in minutes. No infrastructure to provision — Bamboozle handles the control plane, storage, and networking.
-
Create Cluster
In Cloud Control, name your cluster, choose your Kubernetes version, and select a flavor for the master node. Enable HA control plane and integrated monitoring if needed.
-
Add Worker Groups
Add one or more worker groups — each with a separate VM flavor and independent autoscaling configuration. Mix General Compute, Compute Optimised, or High Frequency nodes.
-
Download kubeconfig
When the cluster is ready, click Kubernetes Access in Cloud Control to download your kubeconfig. Connect kubectl, Helm, ArgoCD, or any Kubernetes toolchain immediately.
-
Deploy Workloads
Deploy using standard Kubernetes manifests, Helm charts, or GitOps pipelines. Create PersistentVolumeClaims for NVMe-backed storage, LoadBalancer services with Floating IPs.
-
Autoscale & Monitor
The Cluster Autoscaler handles node scaling automatically as your workloads grow. Prometheus and Grafana provide real-time cluster observability — alerts route to your team.
Ready to Launch Your First Cluster?
Kubernetes clusters on Bamboozle Cloud Engine are self-service and ready in minutes. Start in Cloud Control or contact us to scope the right configuration for your workloads.
FAQ