System Design & Scaling | Forever 32

Forever 32 is an e-commerce clothing website. For this project I inherited front-end work and was assigned to building an optimized back-end that is able to handle high web traffic (minimum requirement: 100 RPS with low latency <2000ms and <1% error rate). I worked with three other engineers, each of us focusing on a different component. I was assigned to Product Overview.

Achievements/Optimizations

• Stess tested locally between MongoDB and PostgreSQL to see which program would be faster for the queries of my component.
  • Result: PostgreSQL was 10ms faster than MongoDB.
• Seeded a PostgreSQL database with over 2 million records and consolidating three tables to optimize the queries.
• Implemented horizontally scaling by making 4 AWS EC2 instances and adding Nginx load balancer (least conn) which resulted in increasing the server to handle 1000 clients per second to 10k clients per second.

Technologies/Tools

• ReactJS
• Node.JS
• Express
• PostgreSQL
• AWS EC2
• Loader.io
• New Relic
• K6
• Nginx

Image of the Front-End

The image's data below is being generated by the my database records.

Developement

npm install
npm run build
npm run start-p
nodemon client/server/index.js