⭐ E-Commerce Ratings & Reviews Service

For PROJECT ATELIER, my team was tasked with replacing the API for an e-commerce clothing retailer using a scalable microservice architecture in order to handle increased web traffic. Each teammate focused on a different microservice. I worked on the backend for the Ratings & Reviews service.

🏆 Achievements and Optimizations

• Seeded a PostgreSQL database with over 20 million records
• Optimized raw SQL query execution times to average < 1ms per query using indexes, joins and PostgreSQL JSON functions
• Improved throughput from around 200 requests per second to over 5000 GET requests per second with <20 ms latency and 0% error rate on average by horizontal scaling between two EC2 server instances with NGINX
• Deployed database and each server using Docker containers for version control

👤 Usage

• Install dependencies using NPM

npm install

• Start server

npm start

• Change the connection string in the database file to refer to your Postgres database
• Run the ETL code to seed your database

🛠️ Technologies

• Node
• Express
• PostgreSQL
• Docker
• Loader.io
• K6
• New Relic
• NGINX

🤝 Authors

• @cry-stal-lee
• @nickwai1
• @jleiandy