🚀 FastAPI Authentication System (Backend Project)

📌 Overview

This project is a production-style FastAPI authentication system designed with scalability, performance, and real-world backend engineering practices in mind.

It demonstrates how to build and optimize a backend system using:

• FastAPI
• PostgreSQL
• Redis
• JWT Authentication
• Load Testing (Locust)

---

🧠 Key Features

• 🔐 JWT-based Authentication (Login/Register)
• 👤 User Management APIs
• ⚡ Redis Caching for performance optimization
• 🗄️ PostgreSQL with optimized connection pooling
• 🧪 Load testing with Locust
• 📊 Performance benchmarking & analysis
• 🔄 Multi-worker scaling using Uvicorn

---

🏗️ Tech Stack

Layer	Technology
Backend	FastAPI
Database	PostgreSQL
Cache	Redis
ORM	SQLAlchemy
Auth	JWT (python-jose)
Password Hashing	bcrypt (passlib)
Load Testing	Locust

---

⚙️ System Architecture

Client → FastAPI → Redis (cache) → PostgreSQL
                ↓
             JWT Auth

---

Architectural Design

Shows interactions between FastAPI, Redis, and PostgreSQL with JWT-based authentication.

🔥 Performance Journey

Phase 1: Initial System

• High latency
• DB bottleneck
• Poor scalability

Phase 2: PostgreSQL Optimization

• Connection pooling introduced
• Improved stability

Phase 3: bcrypt Optimization

• Reduced hashing cost
• Improved response time

Phase 4: Redis Integration

• Cached /users/me
• Eliminated repeated DB reads

---

📊 Benchmark Results

✅ With Redis (300 Users)

• Throughput: ~155 RPS
• Avg Latency: ~456 ms
• Failure Rate: ~0.09%

⚠️ With Redis (500 Users)

• Throughput: ~93 RPS
• Avg Latency: ~1.7 sec
• Failure Rate: ~3%

---

🧠 Key Insights

• Redis removed the database bottleneck
• System shifted from I/O-bound → CPU-bound
• bcrypt hashing became the main limitation
• Worker saturation affects performance under high load

---

📁 Project Structure

FastAPI-Authentication-System/
│
├── .github/
│ └── workflows/
│ └── ci.yml # CI/CD pipeline (GitHub Actions)
│
├── app/
│ ├── core/
│ │ └── redis.py # Redis client setup
│ │
│ ├── routes/
│ │ ├── auth_routes.py # Authentication endpoints (login/register)
│ │ ├── user_routes.py # User endpoints (/users/me)
│ │ ├── redis_routes.py # Redis test/debug endpoints
│ │ └── init.py
│ │
│ ├── init.py
│ ├── auth.py # JWT + password hashing (bcrypt)
│ ├── config.py # Environment configuration
│ ├── database.py # DB connection & session
│ ├── dependencies.py # Auth dependency (get_current_user)
│ ├── logger.py # Logging setup
│ ├── main.py # FastAPI entry point
│ ├── models.py # SQLAlchemy models
│ └── schemas.py # Pydantic schemas
│
├── diagrams/ # System design diagrams
│ ├── architectural_design.png
│ ├── component.png
│ ├── sequence_login.png
│ ├── sequence_users_me.png
│ ├── state_auth.png
│ ├── deployment.png
│ └── data_flow.png
│
├── performance/ # Load testing & benchmarking
│ ├── screenshots/ # Locust UI screenshots
│ ├── results/ # Metrics per phase
│ └── notes/ # Observations & insights
│
├── tests/ # Unit & integration tests
├── logs/ # Application logs
│
├── requirements.txt # Dependencies
├── README.md # Project documentation
├── LICENSE
├── .gitignore
├── test_auth.db # SQLite test DB (local/testing)
└── .test_auth.db # Temporary test DB (CI)

---

🚀 How to Run

1. Clone Repository

git clone <your-repo-url>
cd FastAPI-Authentication-System

2. Install Dependencies

pip install -r requirements.txt

3. Start PostgreSQL

Make sure PostgreSQL is running and configured.

4. Start Redis (Docker)

docker run -d -p 6379:6379 redis

5. Run Server

uvicorn app.main:app --workers 4

---

🧪 Load Testing

Run Locust:

locust -f locustfile.py --host=http://127.0.0.1:8000

Open:

http://localhost:8089

---

🔮 Future Improvements

• Async DB (asyncpg)
• Background password hashing
• Load balancer (Nginx)
• Horizontal scaling (multiple instances)
• Rate limiting using Redis

---

🎯 What This Project Demonstrates

• Backend system design
• Performance optimization
• Load testing & benchmarking
• Bottleneck identification
• Real-world scalability challenges

---

👨‍💻 Author

Aniket Paswan

Aspiring AI/ML Engineer,Backend Engineer

---

⭐ Final Note

This project reflects a real engineering journey:

Database Bottleneck → Redis Optimization → CPU Bottleneck

Understanding this transition is key to designing scalable backend systems.