---
title: "Deploy hydradb"
description: "ast graph database on object storage"
category: "Storage"
url: https://railway.com/deploy/hydradb
---

# Deploy hydradb

ast graph database on object storage

**[Deploy hydradb on Railway](https://railway.com/template/hydradb)**

- **Creator:** kmaki's Projects
- **Category:** Storage

## Template content

### app

- **Image:** xiaosong233/hydradb-railway:latest
- **Start command:** `sh -c 'umask 077; mkdir -p /data/store /data/cache; printf "%s\n" "$GRAPH_AUTH_TOKEN" > /data/auth-token; exec /usr/local/bin/graph-node'`
- **Public domain:** Yes

## Documentation

# Deploy and Host HydraDB on Railway

HydraDB is a Rust graph database built on SlateDB for durable OpenCypher workloads. It provides Neo4j-compatible Bolt access and an authenticated HTTP query API, while using object storage as the source of truth and local disk as cache. This template runs one persistent graph node suitable for evaluation and small single-node deployments.

## About Hosting HydraDB on Railway
Railway runs the pre-built HydraDB container as one service, routes its HTTP query API through a generated HTTPS domain, and attaches a persistent volume for graph data and cache files. The deployment uses plaintext only inside the container because Railway terminates public TLS at the edge.

## Tech Stack
- Rust graph-node runtime
- SlateDB durable storage engine
- OpenCypher query language
- SuiteSparse GraphBLAS traversal kernels
- Bolt 5.1-5.4 and HTTP APIs
- Railway persistent volumes and HTTPS routing

## Why Deploy HydraDB on Railway
Railway provides image-based deployment, managed HTTPS, generated secrets, restart policies, logs, and persistent storage without requiring Kubernetes or a separately managed object store. The template exposes the HTTP API immediately and keeps the database state across restarts.

## Common Use Cases
- Prototype graph-backed applications with OpenCypher
- Evaluate HydraDB through its HTTP query API
- Connect Neo4j-compatible clients over private networking
- Test relationship traversals and graph data models
- Run a compact, persistent single-node graph service

## Deployment Notes
The public HTTP API listens on port 8443; GET /healthz is the browser-friendly health endpoint. Submit queries to /v1/graphs/default/query with Authorization: Bearer  and X-Graph-Namespace: default. HydraDB uses a bearer token and does not define a username/password login. The startup command only writes the generated token to HydraDB's required token file and initializes /data; it then executes the upstream graph-node entrypoint. The /data volume persists the object store and cache. Bolt listens on 7687 and the admin listener on 9090, but only the HTTP query API is publicly routed by this template.

## Dependencies for HydraDB on Railway
HydraDB is self-contained in one application service. Its SlateDB object store and disk cache both use the attached Railway volume, so no Postgres, Redis, or external object-store service is required.

### Deployment Dependencies

| Service | Image | Port | Volume |
|---------|-------|------|--------|
| app | xiaosong233/hydradb-railway:latest | 8443 | /data |

## Similar templates

- [Garage S3 Storage](https://railway.com/deploy/garage-s3-storage) — Ultra-light S3 server: fast, open-source, plug-and-play.
- [Redis](https://railway.com/deploy/redis-1) — Self Host Latest Redis with Railway
- [EasyImg](https://railway.com/deploy/easyimg) — Simple self-hostable Nuxt.js personal image hosting system.

Open this page in a browser: https://railway.com/deploy/hydradb