Skip to main content
Version: v1.0.0-beta

Architecture

Overview of SDN Launch Control architecture and system design.

System Overview​

SDN Launch Control is a monorepo containing:

  • Backend: Django REST API with WebSocket support
  • Frontend: Next.js React application
  • Database: PostgreSQL with TimescaleDB
  • Message Broker: Redis for Celery and WebSockets

Architecture Diagram​

┌─────────────────┐
│   Frontend UI   │ (Next.js + React)
│   (Port 3000)   │
└────────┬────────┘
         │ HTTP/WebSocket
         │
┌────────▼────────┐
│  Backend API    │ (Django + DRF)
│   (Port 8000)   │
└────────┬────────┘
         │
    ┌────┴────┐
    │         │
┌───▼───┐ ┌──▼────┐
│  PostgreSQL │ │  Redis   │
│ TimescaleDB │ │ (Celery) │
└─────────────┘ └──────────┘

Backend Architecture​

Django Applications​

  • network_device: Device management
  • controller: SDN controller management
  • odl: OpenDaylight integration
  • onos: ONOS integration
  • ovs_install: Open vSwitch installation
  • ovs_management: Open vSwitch management
  • classifier: ML-based traffic classification
  • network_data: Network analytics
  • device_monitoring: Real-time statistics
  • account: User authentication
  • notification: Telegram notifications
  • software_plugin: Plugin system
  • ansible: Automation playbooks

Key Technologies​

  • Django 5.0.1: Web framework
  • Django REST Framework: API endpoints
  • Django Channels: WebSocket support
  • Celery: Background tasks
  • Ansible: Network automation
  • TensorFlow/Keras: ML models

Frontend Architecture​

Next.js App Router​

  • Server Components: Default (RSC: true)
  • Client Components: For interactivity
  • API Integration: Axios instance
  • WebSocket: MultiWebSocketContext
  • State Management: React Context API

Key Technologies​

  • Next.js 15.4.2: React framework
  • React 19.1.0: UI library
  • TypeScript 5: Type safety
  • shadcn/ui: UI components
  • Tailwind CSS: Styling
  • Recharts: Data visualization

Data Flow​

Real-time Monitoring​

  1. Network devices send statistics
  2. Backend processes and stores in TimescaleDB
  3. WebSocket pushes updates to frontend
  4. Frontend displays real-time charts

Traffic Classification​

  1. Network flows captured
  2. ML model classifies traffic
  3. Results stored in database
  4. Frontend displays classifications

Database Schema​

TimescaleDB Hypertables​

  • network_data_flow: Flow data
  • network_data_flowstat: Flow statistics

Continuous Aggregates​

  • 1-minute aggregated views
  • Automatic refresh policies
  • Compression enabled

Security​

  • Token-based authentication (Django REST Knox)
  • CORS configuration
  • Environment variable secrets
  • CSRF protection

Scalability​

  • Database connection pooling
  • Celery for async tasks
  • Redis for caching
  • TimescaleDB for time-series optimization

Next Steps​