System requirements

Functional Requirements:

• Create a short URL given a long URL.
• Redirect to the original long URL when the short URL is accessed.

Non-Functional Requirements:

• Availability: The service should be highly available, especially for redirection.
• Response Time: Redirection should take less than 10 ms, while creating a short URL can take up to 10 seconds.
• Scalability: The system should be able to scale with increasing usage.

Capacity estimation

• Requests per second:
  • Create Short URL: 200 req/sec.
  • Redirect URL: 20,000 req/sec.
• Daily Data Generation:
  • Creating Short URLs: 4.4 GB/day.
  • Redirecting URLs: 221 GB/day.

API design

• shortenURL(long_url: str, user_id: str, expiration_time: Optional[datetime]) -> str:
  • Accepts a long URL and returns a short URL.
• redirectURL(short_url: str) -> str:
  • Accepts a short URL and returns the corresponding long URL.

Database design

• Data model:
  • Short URL (Primary Key)
  • Long URL
  • Created time
  • Created by user ID
  • Expiration time

High-level design

The architecture can consist of the following components:

• API Gateway
• Shortening Service: Handles URL creation.
• Mapping Service: Manages URL redirections.
• Database: Stores mappings.
• Cache (e.g., Redis): Improves redirection speed.
• CDN: Caches popular URL mappings for faster access.

High-Level Architecture Diagram:

flowchart TD A[Client] -->|shortenURL()| B[API Gateway] B -->|shortenURL()| C[Shortening Service] A -->|redirectURL()| D[API Gateway] D -->|redirectURL()| E[Mapping Service] E -->|Fetch from Cache| F[Redis Cache] F -->|Cache Hit| G[Redirect to Long URL] E -->|Fetch from DB| H[Database]

There was an issue generating the diagram for the code above.

Request flows

• shortenURL()
  • Client sends request to API Gateway, which forwards it to the Shortening Service.
  • The Shortening Service creates a mapping and stores it in the database.
• redirectURL()
  • The client sends the request to the API Gateway.
  • The Mapping Service checks the cache. If found, it returns the long URL.
  • If not found, it fetches the long URL from the database.

Detailed component design

• Caching:
  • Use Redis for in-memory caching for frequent queries to improve response times.
• Message Queue:
  • For creating URLs, implement a message queue to handle spikes in traffic or delays in database writes.

Trade offs/Tech choices

Explain any trade offs you have made and why you made certain tech choices...

Failure scenarios/bottlenecks

Try to discuss as many failure scenarios/bottlenecks as possible.

Future improvements

• Support for custom short URLs.
• Implement analytics for tracking URL usage (clicks, etc.).