Deploy StarRocks Quickly with Docker and Kubernetes: Step‑by‑Step Guide

What is Cloud‑Native StarRocks?

StarRocks is an MPP analytical database that, since version 3.0, supports a cloud‑native storage‑separation architecture. Deploying it in cloud environments brings elasticity, portability, high reliability, and automated management, which significantly improve cost efficiency and development speed.

StarRocks Architecture Overview

The system consists of a Frontend (FE) node that handles metadata, client connections, query planning and scheduling, and one or more Backend (BE) nodes that store data and execute SQL.

Prerequisites

Docker (latest version)

MySQL client (to connect to StarRocks)

A Kubernetes cluster (Docker Desktop with Kubernetes, Minikube, or any managed K8s service such as Alibaba Cloud ACK, Tencent Cloud TKE, AWS EKS, GCP GKE)

1. Quick Test with Docker

For fast functional verification, run a single‑node StarRocks container that includes one FE and one BE. This setup is intended for demos or small‑data tests only.

docker run -p 9030:9030 -p 8030:8030 -p 8040:8040 -itd starrocks/allin1-ubuntu:latest

After the container starts, wait about a minute, then connect using the MySQL client:

mysql -h127.0.0.1 -P9030 -uroot

2. Deploy with the StarRocks Kubernetes Operator

Install the Operator

Apply the CustomResourceDefinition (CRD) that defines a StarRocksCluster:

kubectl apply -f https://raw.githubusercontent.com/StarRocks/starrocks-kubernetes-operator/main/deploy/starrocks.com_starrocksclusters.yaml

Then install the Operator itself in the default starrocks namespace (or a custom namespace of your choice):

kubectl apply -f https://raw.githubusercontent.com/StarRocks/starrocks-kubernetes-operator/main/deploy/operator.yaml

Deploy a Cluster

Create a YAML manifest that describes the desired number of FE, BE, and CN (compute) nodes. The repository provides a sample starrocks-fe-and-be.yaml which launches three FE and three BE by default; you can edit it to set replication_num=1 and adjust node counts.

wget https://raw.githubusercontent.com/StarRocks/starrocks-kubernetes-operator/main/examples/starrocks/starrocks-fe-and-be.yaml
kubectl apply -f starrocks-fe-and-be.yaml

Connect to the Cluster

Inside the Kubernetes network, obtain the FE service IP: kubectl get svc -n starrocks The service name follows the pattern <cluster-name>-fe-service. Use the IP to connect: mysql -h <FE-IP> -P9030 -uroot For external access, change the FE service type to LoadBalancer or NodePort:

kubectl -n starrocks patch --type=merge src starrockscluster-sample -p '{"spec":{"starRocksFeSpec":{"service":{"type": "LoadBalancer"}}}}'

After the service obtains an ExternalIP, connect from outside the cluster using the same MySQL command.

Reference Links

Docker installation guide: https://docs.docker.com/engine/install/

StarRocks Docker Hub tags: https://hub.docker.com/r/starrocks/allin1-ubuntu/tags

StarRocks Kubernetes Operator repository: https://github.com/StarRocks/starrocks-kubernetes-operator

Operator CRD and manifest files: https://raw.githubusercontent.com/StarRocks/starrocks-kubernetes-operator/main/deploy/starrocks.com_starrocksclusters.yaml and https://raw.githubusercontent.com/StarRocks/starrocks-kubernetes-operator/main/deploy/operator.yaml

Example YAML templates: https://github.com/StarRocks/starrocks-kubernetes-operator/tree/main/examples/starrocks

Following these steps lets you quickly verify StarRocks functionality with Docker and scale to a production‑grade, cloud‑native deployment using Kubernetes.