🚀 Notiflux - Distributed-Ready Notification Engine (Kafka + Spring Boot)

A production-grade, event-driven notification system built using Spring Boot and Apache Kafka. This project demonstrates scalable backend architecture with asynchronous processing, fault tolerance, and clean design patterns.

---

📌 Overview

This system allows users to send notifications (Email, SMS) via a REST API. Instead of processing requests synchronously, messages are pushed to Kafka and handled asynchronously by consumers.

---

🏗️ Architecture

Client → REST API → Kafka (Producer) → Consumer → Strategy → Notification Sent
                                      ↓
                                   DLQ (on failure)

---

⚙️ Tech Stack

• Backend: Spring Boot (Java 21)
• Messaging: Apache Kafka
• Containerization: Docker
• Build Tool: Maven
• Design Patterns: Strategy Pattern
• Architecture Style: Event-Driven Microservice

---

✨ Features

✅ Asynchronous Processing

• Decouples request handling using Kafka
• Improves scalability and responsiveness

✅ Strategy Pattern

• Clean separation for notification channels
• Easily extendable (Email, SMS, Push, etc.)

✅ Retry + Dead Letter Queue (DLQ)

• Failed messages are retried automatically
• After retries, messages are routed to a DLQ topic

✅ Rate Limiting

• Prevents users from spamming the system
• Fixed window rate limiting per user

✅ Fault Tolerance

• Ensures no message loss
• Handles failures gracefully

---

📂 Project Structure

com.example.notificationengine
├── controller/
├── model/
├── service/
├── strategy/
├── config/

---

🚀 Getting Started

1️⃣ Clone the Repository

git clone https://github.com/GeekyHim/Distributed-Notification-Service-Project.git
cd notification-engine

---

2️⃣ Start Kafka using Docker

docker-compose up -d

---

3️⃣ Run the Application

mvn spring-boot:run

---

📩 API Usage

🔹 Send Notification

POST /api/v1/notifications/send

Request Body:

{
  "userId": "123",
  "message": "Hello from Kafka",
  "channel": "EMAIL",
  "recipient": "test@example.com"
}

Note: To simulate a failed respnse, use the word "fail" in message field.

---

🧪 Testing Scenarios

✅ Success Case

• Message processed and sent via selected strategy

❌ Failure Case

• Message retried 3 times
• Then moved to notification-dlq topic

---

📊 Key Concepts Demonstrated

• Event-driven architecture
• Producer-consumer pattern
• Backpressure handling
• Fault tolerance with DLQ
• Rate limiting
• Clean code principles

---

🚀 Future Improvements

• Redis-based distributed rate limiting
• Notification history with database
• API Gateway integration
• Monitoring & logging

---

👨‍💻 Author

Himanshu Gupta
Computer Science Engineering | Chandigarh University

---

⭐ If you found this useful, consider giving it a star!