Designing WhatsApp - Key Points from Lecture

Key Features of WhatsApp

• Group messaging
• Read receipts (Sent, Delivered, and Read ticks)

Important Features to Discuss in System Design Interviews

• Group messaging: Limited to 200 people per group.
• Image/Video sharing: Commonly required.
• Read receipts: Sent, Delivered, and Read statuses.

Additional Features

• Online status: Indicating if a user is online or last seen time.
• Temporary/Permanent chats: Privacy and storage concerns.

Prior Work on Image Sharing

• Reference Tinder video for storing and retrieving images.

Feature Breakdown

1. One-to-One Messaging

• Connect to gateway:
  • Gateway handles external protocol communication.
  • Internally uses efficient protocols without extensive headers.
• Message Routing:
  • TCP connections with minimal state (dumb connections).
  • Sessions microservice maintains user-to-box mapping.
  • WebSockets for real-time communication.
• Message tracking:
  • Sent receipt once message is stored and acknowledged by the server.
  • Delivery receipt upon actual delivery.
  • Read receipt upon message being read by the recipient.
• Mechanisms used:
  • HTTP for message polling (less efficient).
  • WebSockets preferred for real-time chat.

2. Online Status and Last Seen

• Tracking:
  • Server maintains last activity timestamp for users.
  • Differentiates between user activities and app-generated activities.
• Service Logic:
  • Last Seen microservice updates timestamps based on user activities.
  • B queries server for A’s last seen status.
• Online indication:
  • Shows “online” if activity occurred within a recent threshold.
  • Otherwise shows last seen timestamp.

Key Components and Concepts

• Load Balancing: Ensure efficient distribution of messages.
• Service Discovery: Identifies available microservices.
• Heartbeat Mechanism: Maintains active connection status.
• Authentication: Separate service for validating user identity.

3. Group Messaging

• Session Service:
  • Stores user-to-connection mapping.
• Group Service:
  • Manages group membership details.
  • Session service queries group service to get user details.
• Message Routing:
  • Limits group size to reduce system load (e.g., 200 users).
  • Uses consistent hashing for efficient routing.
• Reducing Gateway Load:
  • Offloading parsing tasks to a dedicated parser microservice.
• Ensuring Delivery:
  • Message queuing for retries.
  • Immediate acknowledgment for message receipt.
• Optimization in High Load:
  • Deprioritizing non-essential features (e.g., last seen, delivery receipts) during high load times.

Performance Tips and Techniques

• Graceful degradation: Prioritize important messages under heavy load.
• Consistent Hashing: Efficiently distributes data across servers.
• Message Queuing: Ensures reliable message delivery even on retries.
• WebSockets: Necessary for real-time communication.

Recap

• Focused on four key features: One-to-One messaging, Read receipts, Online status, and Group messaging.
• Emphasized scalability, efficiency, and real-time performance.

Conclusion

• This architecture ensures a resilient, efficient, and scalable chat application.
• Future topics: Load balancing, service discovery, heartbeat maintenance, and authentication.

Thank you for watching!