TicketStorm API

A high-concurrency event ticketing backend built with Django, focused on data integrity and specialized concurrency control.

Overview

The primary challenge of this system is managing high-volume, simultaneous booking requests for limited inventory. This project implements a robust solution to the race condition problem using database-level locking and asynchronous processing.

Technical Implementation

Concurrency and Data Integrity

• Row-Level Locking: Uses PostgreSQL SELECT FOR UPDATE within atomic transactions to prevent double-booking. This ensures that only one request can modify a specific seat status at any given millisecond.
• Idempotency: Implementation of unique idempotency keys to prevent duplicate transactions caused by network retries or client-side double-submits.
• Database Constraints: Utilization of UniqueConstraint and CheckConstraint at the schema level as a final line of defense for data consistency.

Asynchronous Pipeline

• Task Offloading: Heavy operations such as ticket PDF generation and email notifications are handled by Celery workers via a Redis broker.
• Inventory Management: Automated cleanup of expired "Pending" reservations through periodic Celery tasks, ensuring seats return to the available pool if payment is not confirmed within the TTL (Time-To-Live).

System Architecture

• Backend: Python 3.12 / Django 5.0 / Django REST Framework
• Database: PostgreSQL 16
• Task Queue: Redis 7 / Celery 5.4
• Infrastructure: Docker / Docker Compose

Performance Metrics

A load test was conducted simulating 10 concurrent requests targeting a single available seat.

Metric	Value
Concurrent Requests	10
Available Inventory	1
Success Rate	10% (1/10)
Collision Rate	0%
Average Latency	125ms

Setup and Usage

1. Environment Setup:

   docker-compose up -d

2. Initialize Database:

   docker-compose exec web python manage.py makemigrations
   docker-compose exec web python manage.py migrate
   docker-compose exec web python manage.py seed_data

3. Verify Concurrency Logic: A simulation script is provided to demonstrate the locking mechanism under load:

   python3 demo_race.py

4. API Documentation: The OpenAPI schema is available at /api/docs/ (Swagger UI) or /api/redoc/ once the server is running.

Engineering Focus

This project serves as a demonstration of handling:

• Race conditions in distributed systems.
• Pessimistic concurrency control.
• Background job processing and task scheduling.
• Containerized multi-service architecture.