How leberknecht/neuralcoin works
Main Components
a. Data Collection Layer:
• NodeJS Scrapers (for Poloniex, Bittrex, Bitstamp, Bitfinex)
• WebSocket connections to exchanges
• Order book collectors
b. Message Queue Layer:
• RabbitMQ for inter-service communication
• Multiple queues for different purposes (trades, predictions, training)
c. Core Application Layer:
• Symfony 3 Application
• Database (MySQL)
• Entity management (Doctrine)
• WebSocket server for frontend communication
d. Machine Learning Layer:
• Python services
• TensorFlow implementation
• Network creation and training
• Prediction generation
e. Frontend Layer:
• Web interface
• ReactJS components
• Real-time trading tools
• Network visualization
[External Exchanges] -> [Scrapers Layer] -> [Message Queue] -> [Core Application] -> [Frontend]
• ↕
• [Machine Learning Layer]
Detailed Component Relationships
a. Data Flow:
• Trade data flow from exchanges through scrapers
• WebSocket connections for real-time data
• Persistent storage points (MySQL)
• Prediction request/response paths
b. Service Communication:
• RabbitMQ message paths between components
• WebSocket communication to frontend
• File system interactions for network configs
• S3 storage integration points
Key Technical Elements to Highlight
• Docker containers and their relationships
• Database connections and migrations
• Network training data flow
• Real-time prediction pipeline
• Frontend data updates
Infrastructure Components
• Docker containers (show container boundaries)
• Shared volumes
• Network connections
• Cache layers
User Interface (right)
• Container boundaries
• Data flow directions
• Component types
Keep the abstraction level high enough to show the overall architecture while including critical implementation details that affect the system's behavior, such as the queue-based communication and the separation between network creation and training processes.
Remember to indicate the technology stack used in each component:
• NodeJS for scrapers
• Symfony/PHP for core application
• Python/TensorFlow for ML
• ReactJS for frontend
• RabbitMQ for messaging
• MySQL for storage