Intermediate -- 45 min

Deploy to Kubernetes

Install QBITEL Bridge on a Kubernetes cluster using Helm. Configure resource limits, enable observability, and verify all components are healthy and ready for production traffic.

Prerequisites

  • A running Kubernetes cluster (v1.28 or later) -- EKS, GKE, AKS, or local (kind, minikube)
  • Helm 3.12+ and kubectl installed and configured
  • At least 4 CPU cores and 8 GB RAM available in your cluster

What Gets Deployed

Control Plane

  • AI Engine (Python) -- 2 replicas
  • Control Plane (Go) -- 2 replicas
  • Management API (Go) -- 2 replicas

Data Plane

  • Data Plane DaemonSet (Rust) -- 1 per node
  • TimescaleDB -- 1 replica (StatefulSet)
  • Redis Cache -- 1 replica
1

Add the Helm Repository

Add the QBITEL Helm chart repository and update your local chart cache. 

# Add the QBITEL Helm repo
helm repo add qbitel https://charts.qbitel.com
helm repo update

# Search for available charts
helm search repo qbitel

# Output
NAME                    CHART VERSION   APP VERSION   DESCRIPTION
qbitel/qbitel-bridge    1.0.0           1.0.0         QBITEL Bridge - AI Protocol Intelligence
2

Create a Values Override

Create a values-production.yaml file to customize the deployment for your environment. This sets resource limits, enables PQC, and configures persistence. 

# values-production.yaml
global:
  environment: production
  tls:
    enabled: true
    minVersion: "TLSv1.3"

aiEngine:
  replicas: 2
  resources:
    requests:
      cpu: "500m"
      memory: "1Gi"
    limits:
      cpu: "2"
      memory: "4Gi"
  pqc:
    enabled: true
    profile: hybrid

dataplane:
  resources:
    requests:
      cpu: "1"
      memory: "2Gi"
    limits:
      cpu: "4"
      memory: "8Gi"

timescaledb:
  persistence:
    enabled: true
    size: 50Gi
    storageClass: gp3

monitoring:
  prometheus:
    enabled: true
  grafana:
    enabled: true
  opentelemetry:
    enabled: true
    endpoint: "otel-collector:4317"

ingress:
  enabled: true
  className: nginx
  hosts:
    - host: qbitel.your-domain.com
      paths:
        - path: /
          pathType: Prefix
3

Install with Helm

Create the namespace and install the chart with your production values. 

# Create namespace
kubectl create namespace qbitel-system

# Create secrets for database credentials
kubectl create secret generic qbitel-db-credentials \
  --namespace qbitel-system \
  --from-literal=username=qbitel \
  --from-literal=password=$(openssl rand -base64 32)

# Install the chart
helm install qbitel-bridge qbitel/qbitel-bridge \
  --namespace qbitel-system \
  --values values-production.yaml \
  --wait --timeout 5m
4

Verify the Deployment

Confirm that all pods are running and passing health checks. 

# Check pod status
kubectl get pods -n qbitel-system

# Expected output
NAME                              READY   STATUS    RESTARTS   AGE
qbitel-ai-engine-6d8f9c7-abc12   1/1     Running   0          2m
qbitel-ai-engine-6d8f9c7-def34   1/1     Running   0          2m
qbitel-controlplane-5b7c8-gh567  1/1     Running   0          2m
qbitel-controlplane-5b7c8-ij890  1/1     Running   0          2m
qbitel-mgmtapi-4a6b7-kl012       1/1     Running   0          2m
qbitel-mgmtapi-4a6b7-mn345       1/1     Running   0          2m
qbitel-dataplane-pq678            1/1     Running   0          2m
qbitel-timescaledb-0              1/1     Running   0          2m
qbitel-redis-0                    1/1     Running   0          2m

# Run the built-in health check
kubectl exec -n qbitel-system deploy/qbitel-ai-engine -- \
  curl -s http://localhost:8000/health | jq

# Check the Kubernetes health endpoints
kubectl get endpoints -n qbitel-system
5

Access the Console and Grafana

Port-forward to the management API and Grafana to verify the installation through the web interfaces. 

# Port-forward to the QBITEL console
kubectl port-forward -n qbitel-system svc/qbitel-mgmtapi 8080:8080 &

# Port-forward to Grafana
kubectl port-forward -n qbitel-system svc/qbitel-grafana 3001:3000 &

# Open in your browser
# Console: http://localhost:8080
# Grafana: http://localhost:3001 (admin / admin)

What's next? With QBITEL Bridge running on Kubernetes, you can scale the AI engine horizontally, enable auto-scaling with HPA, and integrate with your service mesh. Check the deployment docs for advanced configuration.

Previous: Build a Protocol Adapter Next: Compliance Reporting

Ready for Production Kubernetes?

QBITEL Bridge ships production-grade Helm charts with built-in observability, auto-scaling, and high availability.

Quick Start Guide Star on GitHub Enterprise Support