Design a multi-tenant URL Shortening System

Question

Design a multi-tenant url shortening system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.

Codemia · Accepted Answer

- **Functional Requirements:**  
  1. **URL Shortening:** Allow users to input a long URL and generate a unique short URL.  
  2. **Multi-Tenant Support:** Each tenant (user or organization) should have their own namespace for short URLs.  
  3. **Analytics:** Provide click-through statistics for each short URL, including total clicks and geographic distribution.  
  4. **Redirection:** Support fast redirection from short URLs to original long URLs.  
  5. **API Access:** Expose a RESTful API for creating and retrieving short URLs.  
  
- **Non-Functional Requirements:**  
  1. **Scalability:** System must handle up to 10 million requests per day, with the ability to scale horizontally.  
  2. **Availability:** Aim for 99.99% uptime to ensure users can access short URLs at all times.  
  3. **Performance:** URL redirection must occur in under 100ms for 95% of requests.  
  4. **Security:** Implement authentication and authorization to prevent unauthorized access to tenant data.

- **User Base:** Assume 1 million active users, each creating an average of 10 short URLs per month.  
- **Requests:**  
  - Creation Requests: 1 million users * 10 URLs/month = 10 million URLs/month ~ 333,333 URLs/day.  
  - Redirection Requests: Assuming an average short URL gets 20 clicks/day, total clicks = 10 million URLs * 20 clicks = 200 million clicks/day.  
- **Total Requests:** 333,333 (creation) + 200 million (redirection) = 200,333,333 requests/day, or about 2,315 requests/second on average.  
- **Storage Needs:** Assuming each URL entry requires 256 bytes (including metadata), for 10 million URLs, we need approximately 2.5 GB of storage.

- **Component Diagram:**  
  - **Frontend:** Load balancer (e.g., AWS ELB) to distribute requests.  
  - **API Gateway:** (e.g., AWS API Gateway) for routing and authentication.  
  - **Application Servers:** Stateless microservices (e.g., Node.js or Go) to handle URL creation and redirection.  
  - **Database:**  
    - **Primary Database:** NoSQL database (e.g., DynamoDB) for URL storage to allow for high write throughput and scalability.  
    - **Analytics Database:** Time-series database (e.g., InfluxDB) for storing click-through data.  
  - **Cache:** Redis for caching popular short URLs to improve redirection performance.  
  - **Monitoring:** Prometheus and Grafana for tracking performance metrics.

- **URL Table Schema:**  
  - **Table Name:** `urls`  
  - **Columns:**  
    - `id` (Primary Key, string) - Unique short URL identifier.  
    - `long_url` (string) - Original long URL.  
    - `tenant_id` (string) - Identifier for the tenant.  
    - `created_at` (timestamp) - Date and time of creation.  
    - `click_count` (integer) - Number of times the URL has been accessed.  
  
- **Analytics Table Schema:**  
  - **Table Name:** `url_analytics`  
  - **Columns:**  
    - `url_id` (Foreign Key, string) - Linked to `urls.id`.  
    - `timestamp` (timestamp) - Time of the click.  
    - `geo_location` (string) - Geographic location of the user.

- **Consistency vs. Availability (CAP Theorem):**  
  - Choose eventual consistency for analytics data to improve availability and performance.  
  - Use strong consistency for URL creation to ensure that users always retrieve the correct URL.  
  
- **Database Choice:**  
  - Selecting DynamoDB for its scalability and performance, but at the cost of complex querying capabilities compared to a relational database.  
  
- **Caching Strategy:**  
  - Implementing Redis caching for redirections to enhance performance, which may lead to stale data for a short time if the underlying data changes.  
  
- **Deployment Strategy:**  
  - Deploying across multiple regions for low-latency access, but this introduces complexity in data synchronization and consistency management.

Design a multi-tenant URL Shortening System

Twitter/X

What the Interviewer Expects

Key Topics to Cover

How to Approach This

Possible Follow-up Questions

Practice a Similar Problem on Codemia

Sample Answer

Requirements

Capacity Estimation

Submit Your Answer

Twitter/X Machine Learning Engineer Interview Guide

Related Questions

Design a high-throughput Inventory Management System

Design a low-latency Rate Limiting System

Design a fault-tolerant Payment System

Design a fault-tolerant Messaging System

Design Walmart Product Search