One-to-one messaging in HLD - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Architecture

intermediate

Design Components for One-to-One Messaging

Which component is essential to ensure message delivery confirmation in a one-to-one messaging system?

A Load Balancer to distribute incoming messages evenly

B Cache Layer to store user profiles temporarily

C Message Queue to store and forward messages reliably

D Analytics Module to track message read rates

💡 Hint

Think about how to guarantee messages are not lost and are delivered even if the receiver is offline.

A message queue ensures messages are stored and forwarded reliably, confirming delivery even if the receiver is temporarily unavailable.

❓ scaling

intermediate

Scaling User Connections in One-to-One Messaging

What is the best approach to handle millions of simultaneous user connections in a one-to-one messaging system?

A Implement horizontal scaling with multiple servers and sticky sessions

B Use client polling every second to check for new messages

C Store all messages in a single database instance

D Use a single centralized server with multi-threading

💡 Hint

Consider how to distribute load and maintain user session consistency.

Horizontal scaling with multiple servers and sticky sessions allows distributing user connections while keeping session data consistent.

❓ tradeoff

advanced

Tradeoff Between Consistency and Availability

In a one-to-one messaging system, choosing between strong consistency and high availability during network partitions is a challenge. Which option best describes this tradeoff?

A Favor consistency by blocking message delivery until all replicas confirm receipt

B Favor availability by allowing messages to be delivered even if some replicas are out of sync

C Ignore network partitions and continue normal operations

D Use a single database to avoid replication issues

💡 Hint

Think about user experience during network issues and how message delivery should behave.

Favoring availability allows users to send and receive messages even if some data replicas are temporarily inconsistent, improving user experience.

🧠 Conceptual

advanced

Message Ordering Guarantees

Which design choice ensures strict message ordering in a one-to-one messaging system?

A Deliver messages as soon as they arrive without buffering

B Store messages in multiple shards without coordination

C Use eventual consistency for message storage

D Assign a unique, incrementing sequence number to each message per conversation

💡 Hint

Consider how to keep track of the order in which messages are sent and received.

Assigning a unique, incrementing sequence number per conversation allows the system to deliver messages in the exact order they were sent.

❓ estimation

expert

Capacity Estimation for One-to-One Messaging

Estimate the storage needed per day for a one-to-one messaging system with 10 million active users, each sending 50 messages daily. Assume average message size is 1 KB.

A Approximately 500 GB per day

B Approximately 5 TB per day

C Approximately 50 TB per day

D Approximately 500 TB per day

💡 Hint

Calculate total messages and multiply by average message size.

10 million users * 50 messages = 500 million messages daily. 500 million * 1 KB = 500 million KB = ~476.8 GB, which is approximately 500 GB.