Blog Image

System Design Interview Prep

Q1. What is the CAP theorem?

🔹 Question Restated Explain the CAP theorem in distributed systems and its impact on design.

1. Simple metaphor Imagine a bank with 3 branches:

  • Consistency = all branches show same balance instantly.
  • Availability = customers can always withdraw.
  • Partition tolerance = even if one branch loses network, system still works.

You can’t guarantee all three perfectly at once.

2. Deep dive theory

  • CAP theorem: In a distributed system, you can only choose two of:

    • Consistency (every node sees same data at same time).
    • Availability (system responds even during failure).
    • Partition tolerance (works despite network splits).

  • Trade-offs:

    • CP (Consistency + Partition) → e.g., Zookeeper.
    • AP (Availability + Partition) → e.g., DynamoDB, Cassandra.

Q2. What is scalability?

🔹 Question Restated How do you define and achieve scalability in system design?

1. Simple metaphor A restaurant with long queues can either:

  • Add bigger tables (vertical scaling).
  • Add more branches (horizontal scaling).

2. Deep dive theory

  • Vertical scaling: add more CPU, memory.
  • Horizontal scaling: add more servers/nodes.
  • Scalability bottlenecks: DB, caching, network.

Q3. What is load balancing and why is it important?

🔹 Question Restated Explain load balancing and its role in distributed systems.

1. Simple metaphor Think of a traffic cop at an intersection directing cars evenly across multiple roads.

2. Deep dive theory

  • Load balancer distributes incoming traffic across servers.

  • Benefits:

    • Prevents overload.
    • Enables horizontal scaling.
    • Provides failover.
    • Algorithms: Round-robin, least connections, weighted routing.

Q4. What is caching and why is it important?

🔹 Question Restated Why do systems use caching, and what are the strategies?

1. Simple metaphor Instead of asking the chef to cook from scratch every time, keep popular dishes pre-made near the counter.

2. Deep dive theory

  • Caching = storing frequently accessed data in faster storage.

  • Types:

    • Client-side (browser cache).
    • CDN (Cloudflare, Akamai).
    • Server-side (Redis, Memcached).
    • Eviction policies: LRU, LFU, TTL.