Overview

This project is a YouTube clone designed to enhance my portfolio and deepen my understanding of video processing and backend managment. It features a robust backend focused on handling video processing, storage, and serving, with a minimalistic frontend.

Key Features

• Video Publishing: Users can upload videos to be processed and stored.
• Video Watching: Users can watch videos stored on AWS S3 using MPEG-DASH (adaptive bitrate).
• Backend Services: Two servers handling different aspects of the application:
  • TypeScript Server: Manages user creation and raw video uploads.
  • Go Server: Manages comments, likes, and subscriptions.
• Video Processing: Utilizes a microservice architecture with RabbitMQ for uploading. Uses MPEG-DASH and stores videos on s3 with that, also has a sidecar for thumbnail generation.
• Database Management: PostgreSQL with Sequelize ORM on TS and no orm on Go.
• Notification System: Implemented via SMTP.

Technical Stack

• Frontend: Only for testing the streaming.
• Backend:
  • Languages: TypeScript, Go.
  • Database: PostgreSQL.
  • ORM: Sequelize for TS and no orm on Go.
  • Message Broker: RabbitMQ.
• Cloud Services: AWS S3.
• Servers:
  • API Gateway: NGINX.
  • Load balancer: NGINX.
  • Video Processing: Microservice.

File Structure

• conf/: Configuration files for NGINX.
• frontend/: Frontend application files.
• HighQualityMicroservice/: Microservice for video processing.
• logs/: Log files.
• secondServer/: Go server for comments, likes, and subscriptions.
• typescriptServer/: Main TypeScript server for user and video management.

Getting Started

If you want to run it locally first download all of the dependencies. Then set up the .env variables on the 3 main paths. Finally you have to start docker and can run make and the makefile will start

License

MIT