Quick Start

DataMate supports two main deployment methods:

• Docker Compose: Suitable for quick experience and development testing
• Kubernetes/Helm: Suitable for production deployment

Prerequisites

Docker Compose Deployment

• Docker 20.10+
• Docker Compose 2.0+
• At least 4GB RAM
• At least 10GB disk space

Kubernetes Deployment

• Kubernetes 1.20+
• Helm 3.0+
• kubectl configured with cluster connection
• At least 8GB RAM
• At least 20GB disk space

5-Minute Quick Deployment (Docker Compose)

1. Clone the Code

git clone https://github.com/ModelEngine-Group/DataMate.git
cd DataMate

2. Start Services

Use the provided Makefile for one-click deployment:

make install

After running the command, the system will prompt you to select a deployment method:

Choose a deployment method:
1. Docker/Docker-Compose
2. Kubernetes/Helm
Enter choice:

Enter 1 to select Docker Compose deployment.

3. Verify Deployment

After services start, you can access them at:

• Frontend: http://localhost:30000
• API Gateway: http://localhost:8080
• Database: localhost:5432

4. Check Service Status

docker ps

You should see the following containers running:

• datamate-frontend (Frontend service)
• datamate-backend (Backend service)
• datamate-backend-python (Python backend service)
• datamate-gateway (API gateway)
• datamate-database (PostgreSQL database)
• datamate-runtime (Operator runtime)

Optional Components Installation

Install Milvus Vector Database

Milvus is used for vector storage and retrieval in knowledge bases:

make install-milvus

Select Docker Compose deployment method when prompted.

Install Label Studio Annotation Tool

Label Studio is used for data annotation:

make install-label-studio

Access: http://localhost:30001

Default credentials:

• Username: admin@demo.com
• Password: demoadmin

Install MinerU PDF Processing Service

MinerU provides enhanced PDF document processing:

make build-mineru
make install-mineru

Install DeerFlow Service

DeerFlow is used for enhanced workflow orchestration:

make install-deer-flow

Using Local Images for Development

If you’ve modified local code, use local images for deployment:

make build
make install dev=true

Offline Environment Deployment

For offline environments, download all images first:

make download SAVE=true

Images will be saved in the dist/ directory. Load images on the target machine:

make load-images

Uninstall

Uninstall DataMate

make uninstall

The system will prompt whether to delete volumes:

• Select 1: Delete all data (including datasets, configurations, etc.)
• Select 2: Keep volumes

Uninstall Specific Components

# Uninstall Label Studio
make uninstall-label-studio

# Uninstall Milvus
make uninstall-milvus

# Uninstall DeerFlow
make uninstall-deer-flow

Next Steps

• Installation Guide - Detailed installation and configuration
• System Architecture - Understand DataMate architecture
• Development Setup - Local development environment

Common Questions

Q: What if service startup fails?

First check if ports are occupied:

# Check port usage
lsof -i :30000
lsof -i :8080

If ports are occupied, modify port mappings in deployment/docker/datamate/docker-compose.yml.

Q: How to view service logs?

# View all service logs
docker compose -f deployment/docker/datamate/docker-compose.yml logs

# View specific service logs
docker compose -f deployment/docker/datamate/docker-compose.yml logs -f datamate-backend

Q: Where is data stored?

Data is persisted through Docker volumes:

• datamate-dataset-volume: Dataset files
• datamate-postgresql-volume: Database data
• datamate-log-volume: Log files

View all volumes:

docker volume ls | grep datamate