How thykp/esd_team8 works
Main Components
• Frontend: A React-based UI built with Vite, TypeScript, Tailwind CSS, and ShadCN UI. It connects to Supabase for authentication.
• Backend: A microservices architecture consisting of “simple” and “complex” services. These services use different languages – Go, JavaScript/TypeScript, and Python – and frameworks like Gin, Express, Hono, FastAPI, and Flask.
• API Gateway: Kong is used to aggregate and manage requests across microservices.
• Message Broker: RabbitMQ is the primary messaging system enabling real-time communication between services.
• Databases/Storage: The project mentions a mix of SQL and NoSQL databases and uses solutions such as Firebase, Google Cloud Storage, and Supabase for storage and data persistence.
• Orchestration & Deployment: Kubernetes YAML files (deployments, services, configmaps, etc.) indicate that microservices are containerized and deployed on a Kubernetes cluster managed partly via Terraform scripts. There are CI/CD pipelines (GitHub Actions, Snyk, Checkov) in place for automated builds and deployments.
• Monitoring and Logging: Use of Prometheus and Grafana for metrics, along with logging utilities in individual services.
• Infrastructure as Code: Terraform configurations manage aspects such as the GKE cluster, storage, load balancers, and versioning.
Component Relationships and Interactions
• The frontend application sends user interactions and API calls to the backend via REST or gRPC.
• The Kong API Gateway fronts the backend microservices, handling CORS, rate limiting, and custom authentication.
• Backend microservices (classified as either “simple” or “complex”) handle specific business logic (e.g., ordering, payment, menu management, chat, recommendation) and communicate asynchronously via RabbitMQ.
• The microservices interact with their respective databases and external storage systems.
• CI/CD pipelines automate the building, testing, and deployment of Docker images, which are deployed to Kubernetes (using Kubernetes manifests from the “backend/kubernetes” directory).
• Monitoring data flows from microservices to Prometheus and Grafana, making cluster performance visible.
Highlighting Architectural Patterns & Technologies
• Microservices Architecture: The system is modular, with services loosely coupled and dedicated to specific business functions.
• Language/Framework Agnosticism: Different microservices use various technology stacks.
• Containerization and Orchestration: Docker (with docker-compose for local development) and Kubernetes for production deployments.
• Cloud-Native CI/CD: Automated pipelines are in place, including security scanning (Snyk, Checkov).
• Real-time inter-service communication: Implemented via RabbitMQ.
• API Gateway Pattern: Implemented by Kong to centralize API management.
Final Suggestions
• Every component is labeled (e.g., “Frontend – React App,” “Backend Service – Order Processing,” “RabbitMQ,” “Kong Gateway”).