Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/qdrant/qdrant/llms.txt

Use this file to discover all available pages before exploring further.

Qdrant can be deployed on Kubernetes for production workloads with high availability and scalability. This guide covers deployment configurations using StatefulSets for data persistence.

​
Quick Start with Deployment

For a simple stateless deployment (development only): 
apiVersion: apps/v1
kind: Deployment
metadata:
  name: qdrant
  namespace: default
spec:
  replicas: 1
  selector:
    matchLabels:
      app: qdrant
  template:
    metadata:
      labels:
        app: qdrant
    spec:
      containers:
      - name: qdrant
        image: qdrant/qdrant:latest
        ports:
        - containerPort: 6333
          name: http
          protocol: TCP
        - containerPort: 6334
          name: grpc
          protocol: TCP
        env:
        - name: QDRANT__SERVICE__GRPC_PORT
          value: "6334"
        - name: QDRANT__LOG_LEVEL
          value: "INFO"
        livenessProbe:
          httpGet:
            path: /healthz
            port: 6333
          initialDelaySeconds: 30
          periodSeconds: 10
          timeoutSeconds: 5
        readinessProbe:
          httpGet:
            path: /readyz
            port: 6333
          initialDelaySeconds: 10
          periodSeconds: 5
          timeoutSeconds: 3
        resources:
          requests:
            memory: "512Mi"
            cpu: "250m"
          limits:
            memory: "2Gi"
            cpu: "1000m"
This deployment does not persist data. Use StatefulSet for production workloads.

​
StatefulSet for Persistence

For production deployments with persistent storage: 
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: qdrant
  namespace: default
spec:
  serviceName: qdrant-headless
  replicas: 1
  selector:
    matchLabels:
      app: qdrant
  template:
    metadata:
      labels:
        app: qdrant
    spec:
      containers:
      - name: qdrant
        image: qdrant/qdrant:latest
        ports:
        - containerPort: 6333
          name: http
        - containerPort: 6334
          name: grpc
        - containerPort: 6335
          name: p2p
        env:
        - name: QDRANT__SERVICE__GRPC_PORT
          value: "6334"
        - name: QDRANT__CLUSTER__P2P__PORT
          value: "6335"
        volumeMounts:
        - name: qdrant-storage
          mountPath: /qdrant/storage
        - name: qdrant-snapshots
          mountPath: /qdrant/snapshots
        livenessProbe:
          httpGet:
            path: /livez
            port: 6333
          initialDelaySeconds: 30
          periodSeconds: 10
          timeoutSeconds: 5
          failureThreshold: 3
        readinessProbe:
          httpGet:
            path: /readyz
            port: 6333
          initialDelaySeconds: 10
          periodSeconds: 5
          timeoutSeconds: 3
          failureThreshold: 3
        resources:
          requests:
            memory: "2Gi"
            cpu: "500m"
          limits:
            memory: "8Gi"
            cpu: "2000m"
  volumeClaimTemplates:
  - metadata:
      name: qdrant-storage
    spec:
      accessModes: [ "ReadWriteOnce" ]
      storageClassName: standard
      resources:
        requests:
          storage: 20Gi
  - metadata:
      name: qdrant-snapshots
    spec:
      accessModes: [ "ReadWriteOnce" ]
      storageClassName: standard
      resources:
        requests:
          storage: 10Gi

​
Service Configuration

Create services to expose Qdrant:

​
LoadBalancer Service (HTTP/gRPC)

apiVersion: v1
kind: Service
metadata:
  name: qdrant
  namespace: default
spec:
  type: LoadBalancer
  selector:
    app: qdrant
  ports:
  - name: http
    port: 6333
    targetPort: 6333
    protocol: TCP
  - name: grpc
    port: 6334
    targetPort: 6334
    protocol: TCP

​
Headless Service (for StatefulSet)

apiVersion: v1
kind: Service
metadata:
  name: qdrant-headless
  namespace: default
spec:
  clusterIP: None
  selector:
    app: qdrant
  ports:
  - name: http
    port: 6333
    targetPort: 6333
  - name: grpc
    port: 6334
    targetPort: 6334
  - name: p2p
    port: 6335
    targetPort: 6335

​
ClusterIP Service (Internal)

apiVersion: v1
kind: Service
metadata:
  name: qdrant-internal
  namespace: default
spec:
  type: ClusterIP
  selector:
    app: qdrant
  ports:
  - name: http
    port: 6333
    targetPort: 6333
  - name: grpc
    port: 6334
    targetPort: 6334

​
Configuration with ConfigMap

Store configuration in a ConfigMap: 
apiVersion: v1
kind: ConfigMap
metadata:
  name: qdrant-config
  namespace: default
data:
  config.yaml: |
    log_level: INFO
    
    storage:
      storage_path: /qdrant/storage
      snapshots_path: /qdrant/snapshots
      on_disk_payload: true
      
      performance:
        max_search_threads: 0
        optimizer_cpu_budget: 0
    
    service:
      host: 0.0.0.0
      http_port: 6333
      grpc_port: 6334
      enable_cors: true
Mount the ConfigMap in your StatefulSet: 
volumeMounts:
- name: qdrant-config
  mountPath: /qdrant/config
  readOnly: true

volumes:
- name: qdrant-config
  configMap:
    name: qdrant-config

​
Secrets for API Keys

Store sensitive data in Kubernetes Secrets: 
apiVersion: v1
kind: Secret
metadata:
  name: qdrant-secrets
  namespace: default
type: Opaque
stringData:
  api-key: your-secret-api-key-here
  read-only-api-key: your-read-only-key-here
Reference secrets in your deployment: 
env:
- name: QDRANT__SERVICE__API_KEY
  valueFrom:
    secretKeyRef:
      name: qdrant-secrets
      key: api-key
- name: QDRANT__SERVICE__READ_ONLY_API_KEY
  valueFrom:
    secretKeyRef:
      name: qdrant-secrets
      key: read-only-api-key

​
Health Checks

Qdrant provides native Kubernetes health check endpoints: 
livenessProbe:
  httpGet:
    path: /livez
    port: 6333
  initialDelaySeconds: 30
  periodSeconds: 10
  timeoutSeconds: 5
  failureThreshold: 3

​
Resource Recommendations

​
Small Deployment (Development)

resources:
  requests:
    memory: "512Mi"
    cpu: "250m"
  limits:
    memory: "2Gi"
    cpu: "1000m"

​
Medium Deployment (Production)

resources:
  requests:
    memory: "4Gi"
    cpu: "1000m"
  limits:
    memory: "16Gi"
    cpu: "4000m"

​
Large Deployment (High Traffic)

resources:
  requests:
    memory: "16Gi"
    cpu: "4000m"
  limits:
    memory: "64Gi"
    cpu: "16000m"

​
Ingress Configuration

Expose Qdrant via Ingress: 
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: qdrant-ingress
  namespace: default
  annotations:
    nginx.ingress.kubernetes.io/backend-protocol: "HTTP"
    cert-manager.io/cluster-issuer: "letsencrypt-prod"
spec:
  ingressClassName: nginx
  tls:
  - hosts:
    - qdrant.example.com
    secretName: qdrant-tls
  rules:
  - host: qdrant.example.com
    http:
      paths:
      - path: /
        pathType: Prefix
        backend:
          service:
            name: qdrant
            port:
              number: 6333

​
Helm Chart (Community)

While Qdrant doesn’t provide an official Helm chart yet, you can create a basic one or use community-maintained charts: 
# Create a basic Helm chart structure
helm create qdrant-chart

# Edit values.yaml with your configuration
# Install the chart
helm install qdrant ./qdrant-chart -n qdrant --create-namespace
The Qdrant community maintains unofficial Helm charts. Check the Qdrant GitHub organization for the latest options.

​
Horizontal Pod Autoscaling

For cluster deployments, configure HPA: 
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: qdrant-hpa
  namespace: default
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: StatefulSet
    name: qdrant
  minReplicas: 3
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 70
  - type: Resource
    resource:
      name: memory
      target:
        type: Utilization
        averageUtilization: 80
When using HPA with StatefulSets, ensure your cluster is properly configured with replication and sharding.

​
Monitoring

Qdrant exposes Prometheus metrics at /metrics: 
apiVersion: v1
kind: Service
metadata:
  name: qdrant-metrics
  namespace: default
  labels:
    app: qdrant
  annotations:
    prometheus.io/scrape: "true"
    prometheus.io/port: "6333"
    prometheus.io/path: "/metrics"
spec:
  selector:
    app: qdrant
  ports:
  - name: metrics
    port: 6333
    targetPort: 6333

​
Next Steps

Distributed Deployment

Set up a multi-node cluster

Configuration

Tune performance and storage

Security

Enable TLS and authentication

Docker

Docker deployment guide