Apache Druid

Deploy Apache Druid on Kubernetes — a high-performance, real-time analytics database designed for fast slice-and-dice analytics on large datasets, commonly used for BI dashboards and real-time data exploration.

Key Features

6-component architecture — coordinator, overlord, broker, router, historical, middlemanager
Router web console — built-in Druid console accessible via ingress
PostgreSQL metadata store — bundled subchart or external PostgreSQL/MySQL
ZooKeeper coordination — bundled subchart or external ZooKeeper
Deep storage — local or S3-compatible (MinIO, AWS S3)
Persistent volumes — segment cache (historical) and task storage (middlemanager)
Per-component scaling — independent replica counts and JVM tuning
Health probes — liveness on /status/health, readiness on /status/selfDiscovered
Ingress support — configurable class (traefik, nginx, etc.)

Installation

HTTPS repository:

helm repo add helmforge https://repo.helmforge.dev
helm repo update
helm install druid helmforge/druid -f values.yaml

OCI registry:

helm install druid oci://ghcr.io/helmforgedev/helm/druid -f values.yaml

Basic Example

coordinator:
  enabled: true

broker:
  enabled: true

router:
  enabled: true

historical:
  enabled: true
  persistence:
    size: 20Gi

ingress:
  enabled: true
  ingressClassName: traefik
  hosts:
    - host: druid.example.com
      paths:
        - path: /
          pathType: Prefix

S3 Deep Storage

deepStorage:
  type: s3
  s3:
    bucket: my-druid-bucket
    baseKey: druid/segments
    region: us-east-1
    accessKey: 'AKIAIOSFODNN7EXAMPLE'
    secretKey: 'wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY'

External Services

metadata:
  mode: external
  external:
    type: postgresql
    host: postgres.example.internal
    port: 5432
    name: druid
    username: druid
    existingSecret: druid-db

zookeeperConfig:
  mode: external
  external:
    hosts: 'zk1.example.internal:2181,zk2.example.internal:2181'

postgresql:
  enabled: false

zookeeper:
  enabled: false

Key Values

Key	Default	Description
`image.repository`	`apache/druid`	Druid container image
`coordinator.enabled`	`true`	Enable coordinator
`coordinator.javaOpts`	`-Xms256m -Xmx512m`	Coordinator JVM options
`overlord.enabled`	`true`	Enable overlord
`broker.enabled`	`true`	Enable broker
`broker.javaOpts`	`-Xms512m -Xmx1g`	Broker JVM options
`router.enabled`	`true`	Enable router (web console)
`historical.enabled`	`true`	Enable historical node
`historical.persistence.size`	`10Gi`	Segment cache volume size
`middleManager.enabled`	`true`	Enable middle manager
`middleManager.workerCapacity`	`2`	Tasks per middle manager
`metadata.mode`	`subchart`	Metadata mode: subchart or external
`zookeeperConfig.mode`	`subchart`	ZooKeeper mode: subchart or external
`deepStorage.type`	`local`	Deep storage: local or s3
`ingress.enabled`	`false`	Enable ingress

Operational Notes

Druid is memory-intensive — all 6 components together require 2GB+ RAM minimum
tune JVM heap per component via javaOpts for production sizing
the router component serves the web console and is the default ingress target
use S3 deep storage for production; local deep storage is only suitable for development
historical nodes cache segments on disk — size the PVC based on data volume

More Information

See the source code and full values reference on GitHub.