A collection of ready-to-run deployment configurations for ArcadeDB, from single-node development setups to production-grade HA clusters.
| Scenario | Description | Orchestrator |
|---|---|---|
| HA Cluster | 3-node high-availability cluster with Raft consensus | Docker Compose |
| Kubernetes | 3-node HA cluster deployed via Helm on a local kind cluster | Kubernetes / Helm |
| Spring Cluster | 3-node embedded ArcadeDB HA cluster inside Spring Boot apps, serving the recommendation engine over REST | Docker Compose |
Each scenario has its own workflow that runs on changes to that directory. To test every scenario at once, run the All Modules CI workflow from the Actions tab, or with the GitHub CLI:
gh workflow run all-modules.yml- Docker >= 24.0
- Docker Compose >= 2.0 (ha-cluster and spring-cluster scenarios)
- kind >= 0.24.0, Helm >= 3.16,
kubectl(kubernetes scenario) - JDK 25 and Maven 3.9+ (spring-cluster local builds)
curlandjq
Each scenario directory contains:
start.sh— bring the deployment upstop.sh— tear the deployment down (kubernetes scenario)test.sh— verify the deployment is workingREADME.md— scenario-specific details and configuration options