Java Spring Boot Redis Kafka Docker License

A production-ready Flash Sale Engine built with Spring Boot, designed to safely handle massive concurrent traffic and limited-stock purchases.

This project simulates real-world flash sale systems used by Amazon, Flipkart, Myntra, where thousands of users attempt to buy the same product at the same time—without overselling, crashing, or being abused by bots.

• 🛡️ User-based Rate Limiting (anti-bot & abuse protection)
• 🛒 Atomic Purchase Flow (no overselling)
• ⚙️ High-Concurrency Safe Design
• 🚀 Redis-powered performance optimization
• 📦 Kafka-based asynchronous order processing
• 🔐 JWT-secured APIs with Spring Security
• 🧪 RESTful API architecture
• 🐳 Docker-ready deployment
• 📊 Clean & scalable layered architecture

• Applied per user to restrict excessive purchase attempts
• Protects against DDoS, brute-force, and bot traffic
• Implemented using Redis counters for low-latency checks
• Exceeds limit → returns 429 TOO_MANY_REQUESTS

• Guarantees atomic stock decrement
• Prevents race conditions and overselling
• Uses transactional boundaries at service/database level
• Designed to stay consistent under thousands of parallel requests

• Decouples API layer from order processing

• Improves throughput, scalability, and fault tolerance

• Enables future extensions like:

  • Retry queues
  • Dead Letter Queue (DLQ)
  • Event-driven analytics

FlashSaleEngine
├── controller # REST Controllers
├── service # Business Logic
├── repository # Database Access Layer
├── model # Entities & DTOs
├── security # JWT & Security Config
├── config # Application Configurations
├── kafka # Kafka Producer & Consumer
├── redis # Cache & Rate Limiting Logic
└── exception # Global Exception Handling

POST /buy/{productId}

if (!rateLimitingService.tryAccess(user.getId())) {
 return ResponseEntity
 .status(HttpStatus.TOO_MANY_REQUESTS)
 .body("Too many attempts, try again later!");
}

docker build -t flash-sale-engine .
docker run -p 8080:8080 flash-sale-engine

• Java 21
• Maven
• MySQL
• Redis
• Apache Kafka

git clone https://github.com/Ash8389/FlashSaleEngine.git
cd FlashSaleEngine
mvn clean install
mvn spring-boot:run

Simulate real flash sale traffic using JMeter.

• Validate rate limiting
• Ensure zero overselling
• Measure response time under load
• Verify system stability

• Threads: 1000+
• Ramp-up: 10–30 seconds
• Loop Count: 1

Request

• POST /buy/{productId}

• Headers:

  • Authorization: Bearer <JWT_TOKEN>
  • Content-Type: application/json

jmeter -n -t flash-sale-test.jmx -l results.jtl

• ✔️ Successful purchases (200 OK)
• ⛔ Excess traffic blocked (429 TOO_MANY_REQUESTS)
• 🔒 No negative stock values
• ⚡ Stable response times

• Distributed locking with Redisson
• Monitoring with Prometheus & Grafana
• Retry mechanism + DLQ
• Kubernetes deployment
• Advanced stress & chaos testing

Ashish Kumar Jha Backend Developer | Java | Spring Boot | Redis | Kafka

🔗 GitHub: https://github.com/Ash8389

⭐ If you find this project useful, consider starring the repository!