Back to Projects
backendCase Study

Distributed URL Shortening Platform

Built a distributed URL shortening platform with real-time click analytics using microservices and event-driven architecture.

Distributed URL Shortening Platform

Overview

Developed a scalable URL shortening platform composed of multiple microservices responsible for URL management, redirection, analytics processing, and reporting. The system combines low-latency synchronous redirect APIs with asynchronous event-driven analytics pipelines powered by Kafka Streams. Redis caching and PostgreSQL query optimizations enable high-throughput URL resolution, while click events are processed in real time to generate detailed analytics and usage insights.

Key Features

  • 1Built 4+ independently deployable microservices for URL management, redirection, analytics, and reporting
  • 2Implemented low-latency URL redirect APIs with optimized lookup workflows
  • 3Designed event-driven click analytics pipeline using Kafka and Kafka Streams
  • 4Processed click events asynchronously to generate real-time usage statistics and engagement metrics
  • 5Integrated Redis caching layer to accelerate URL resolution and reduce database pressure
  • 6Optimized PostgreSQL indexing and query execution plans for high-volume redirect traffic
  • 7Implemented scalable REST APIs for URL creation, management, and analytics retrieval

The Engineering Challenge

Building a system capable of handling high redirect throughput while simultaneously processing analytics events without impacting user-facing latency. This was solved by separating synchronous redirect workflows from analytics processing using Kafka Streams, allowing click events to be processed asynchronously while maintaining fast redirect response times. Additional performance gains were achieved through Redis caching and PostgreSQL query optimization.

Project Links

View RepositoryLive Demo Unavailable

Technologies

JavaSpring BootKafkaKafka StreamsPostgreSQLRedisDockerJMeterREST APIMicroservices

Key Outcomes

Sustained 1200+ requests per second during load testing
Achieved 45ms p95 latency under stress-test workloads
Significantly reduced database load through Redis caching
Delivered real-time analytics processing using streaming architecture