PAD Laboratory Work nr 1.

Checkpoint 1

Project name - Netflix Clone App

1. Why is necessary for the idea to build a distributed system?

• Scalability – in case if we have large number of users or requests, a distributed system allows to scale different components independently.
• Fault Tolerance – In distributive system, a failure will not affect the entire system.
• Maintainability – we can maintain or update one microservice without impacting any other microservice.

A real world example is Netflix which is based on microservices.

---

2. Define service boundaries

---

3. Technology stack:

For the Content Recommendation Service:

• Node.js (JavaScript):
  • Reasoning:
    • Non-blocking, asynchronous I/O model for handling multiple concurrent requests, crucial for real-time content recommendation.
    • Scalability with low memory usage, ideal for managing a large number of user requests.
    • Rich ecosystem of packages and libraries for content analysis, recommendation algorithms, and database management.
• Express.js (Node.js):
  • Reasoning:
    • Lightweight web application framework for building robust and scalable APIs for the Content Recommendation Service.
    • Allows for easy routing, middleware integration, and handling of HTTP requests.
• MongoDB (NoSQL):
  • Reasoning:
    • NoSQL database for flexibility in handling unstructured or semi-structured data, which aligns well with content recommendation systems.
    • Scalable, high-performance solution for managing large volumes of interaction data. For the User Authentication Service:
• Flask (Python):
  • Reasoning:
    • Clear and concise microframework for building specialized services, making it an excellent choice for user authentication.
    • Provides essential features without the overhead of a full-featured framework, allowing for focused and efficient development.
• PostgreSQL (Relational):
  • Reasoning:
    • Reliable and robust relational database for structured user data management, including profiles, credentials, and authentication tokens.

---

4. Design Data Management:

TO BE MADE - Netflix business logic

Data format that will be managed will be JSON format. User microservice will have the following endpoints:

• Register User
  Endpoint: 'POST/ api/users/register'
  Description: Register a User
  Request body:

{
  "email" : "ciuvak@gmail.com",
  "username": "example_user",
  "password": "secure_password"
}

Response body:

{
  "user_id": "123456789",
  "email": "ciuvak@gmail.com",
  "username": "example_user",
  "message": "User registered successfully."
}

• Login User
  Endpoint: 'POST/ api/users/login'
  Description: User Authentication
  Request body:

{
  "username": "example_user",
  "password": "secure_password"
}

Response body:

{
  "user_id": "123456789",
  "username": "example_user",
  "token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwiaWF0IjoxNTE2MjM5MDIyfQ.SflKxwRJSMeKKF2QT4fwpMeJf36POk6yJV_adQssw5c",
  "message": "Login successful."
}

The Content Recommendation Service will have the following endpoints:

• Get recommendations
  Endpoint: 'GET api/recommendations'
  Description: Get a list of recommendations for the user
  Request body:

  • Authorization: Bearer (for user authentication)

  Response body:
  Status 200 OK with a list of recommended content OR Status 401 Unauthorized if token is invalid

• Get content details Endpoint: 'GET api/content/contentId'
  Description: Get details of a selected content
  Request body:

  • Parameters: content identifier

  Response body:
  Status 200 OK with a list of recommended content OR Status 401 Unauthorized if token is invalid

• Like or Dislike Content Endpoint: 'POST api/content/:contentId'/like
  Description: Allows a user to like or dislike a specific piece of content.
  Request body:

{
  "action": "like"  // or "dislike"
}

Response body: Status 200 OK if the action was successful OR Status 401 Unauthorized if token is invalid OR Status 404 Not Found if content with the specified ID does not exist

---

5. Deployment and Scaling:

To deploy my app I will use Docker.