⚡eMotion

• eMotion is a scalable, event-driven ride-hailing platform built with Node.js, Express.js, MongoDB, WebSockets, and RabbitMQ.
• It supports real time ride management with WebSocket notifications and microservice communication for a smooth user experience.

---

🔗 Architecture Overview

eMotion uses a microservices architecture featuring:

• Nginx as a reverse proxy and load balancer.
• Docker Compose to containerize and decouple services.
• RabbitMQ for reliable message queuing and asynchronous communication.
• WebSocket server for real-time ride event notifications.
• MongoDB to store user, captain, and ride data.
• Multiple microservices for User, Captain, Ride, and Central Websocket based Notification handling.

🔗 Communication Flow

1. Users and captains authenticate via JWT-secured REST endpoints.
2. Ride requests are created and managed via REST APIs.
3. Ride status updates (accept, start, end) are propagated via RabbitMQ events.
4. WebSocket server pushes real-time notifications to users and captains.
5. Nginx routes incoming requests to appropriate microservices.

---

⚙️ Tech Stack

Technology	Purpose
Node.js	Backend runtime
Express.js	REST API framework
MongoDB	Persistent data storage
RabbitMQ	Message queue for microservice communication
WebSocket	Real-time notifications
Nginx	Reverse proxy & load balancing
Docker Compose	Container orchestration
Azure VM	Hosting backend infrastructure
Vercel	Frontend deployment

---

🔗 Features

• ✅ User & Captain registration and JWT authentication
• ✅ REST APIs for ride creation, acceptance, start, and completion
• ✅ Real-time ride event notifications via WebSocket with sub-10ms latency
• ✅ Reliable, decoupled microservices communication using RabbitMQ
• ✅ Scalable multi-instance backend containerized with Docker Compose
• ✅ Nginx reverse proxy managing load balancing between services

---

🔗 Performance Metrics from Load Testing

• Successfully handled 306+ concurrent WebSocket sessions with average connection latency under 10ms
• Supported over 393 ride event checks with 74% success rate at peak load (15 VUs, 8 mins)
• Average HTTP request duration: ~11.5 seconds, median: 1.78 seconds under stress
• Maintained 100% successful user and captain logins during tests
• RabbitMQ ensured reliable asynchronous messaging with minimal overhead and zero message loss
• Dockerized microservices maintained clear separation and easy scaling in Azure VM environment

---

🛠️ Deployment & Setup

• Frontend is currently hosted on https://e-motion-eight.vercel.app
• Backend is currently hosted on https://distros.tech

1. Prerequisites

• Docker & Docker Compose installed
• Azure Virtual Machine for backend deployment
• Nginx installed on VM as reverse proxy
• MongoDB instance (can be hosted or containerized)
• RabbitMQ instance running (can be containerized)
• Vercel account for frontend hosting

2. Clone the repository

git clone https://github.com/alucard017/eMotion-Backend.git
cd eMotion-Backend

git clone https://github.com/alucard017/eMotion-frontend.git
cd eMotion-frontend

3. Configure environment variables

• Create a .env file inside each user, ride, captain services following .env.sample with values like:
• Creating .env is optional if you want to directly provide environments via docker-compose.
• If not using docker-compose you need to run the gateway server or else if using docker-compose you need to run nginx and configure nginx.conf as given as sample.
• docker-compose file sample is also provided to follow the structure strictly.

4. Running services

• For frontend it will be running on localhost:3000 if 3000 port is not available it will run in port 3001 alterntaively.
• Make sure to add .env variable in frontend too and make sure to use the API_BASE_URL of backend to make API calls. Currently frontend assumes the request comes from hosted frontend and backend so API_BASE_URL is ignored.
• For backend if not using docker-compose simply go to each service folders and run.

npm run build
npm run start

• Service will start successfuly.
• Make sure that all services will be connected via gateway server.
• If using docker-compose then just run the below command

docker-compose up --build -d

• You can see the logs by

docker compose logs -f

• Make sure when using docker-compose, place nginx.conf as in the same file where docker-compose lies.