Question 1

What is the difference between Vertical Scaling and Horizontal Scaling?

Accepted Answer

Vertical Scaling (scaling up) means adding more power (CPU, RAM) to an existing server machine. It is simple, has no network overhead, but has hardware limits and introduces a Single Point of Failure (SPOF). Horizontal Scaling (scaling out) means adding more machines/servers to your pool. It supports infinite scaling, provides high resilience, but requires load balancers and introduces distributed network complexity.

Question 2

What is a Load Balancer, and how does it route traffic?

Accepted Answer

A Load Balancer is a device or software service that distributes incoming network traffic across a group of backend servers. This prevents any single server from becoming overloaded. Common routing algorithms include:
1. Round Robin: Sequential distribution.
2. Least Connections: Routes to the server with the fewest active sessions.
3. IP Hash: Routes based on client IP hash, ensuring a client always hits the same server (sticky sessions).

Question 3

Explain the CAP Theorem.

Accepted Answer

The CAP Theorem states that a distributed system can guarantee at most two out of three characteristics simultaneously:
1. Consistency: Every read receives the most recent write or an error.
2. Availability: Every non-failing node returns a response (without guarantee of containing the latest write).
3. Partition Tolerance: The system continues to operate despite network partition drops or message delays.
Since network partitions are inevitable in distributed systems, you must choose between Consistency (CP) or Availability (AP) during a network partition.

Question 4

What is a Content Delivery Network (CDN), and how does it work?

Accepted Answer

A CDN is a geographically distributed network of proxy servers (Edge Locations) that cache static content (HTML, CSS, images, videos) closer to the end user. When a user requests an asset, the request is routed to the nearest Edge Location, minimizing latency and reducing load on the origin server.

Question 5

Explain Database Sharding vs Partitioning.

Accepted Answer

Database Partitioning (vertical or horizontal) divides tables within a *single* database instance to improve performance. Database Sharding is a horizontal partitioning technique where database rows are split and stored across *multiple, separate* physical database servers (shards), allowing databases to scale horizontally.

Question 6

What is the difference between ACID and BASE properties?

Accepted Answer

ACID properties (Atomicity, Consistency, Isolation, Durability) are followed by relational databases (RDBMS) to guarantee strict data integrity, transactions, and immediate consistency. BASE properties (Basically Available, Soft state, Eventual consistency) are followed by NoSQL databases to optimize for high availability and partition tolerance, permitting data to become eventually consistent over time.

Question 7

What is a Message Queue (MQ), and why is it used?

Accepted Answer

A Message Queue (like RabbitMQ, Apache Kafka, or SQS) is a form of asynchronous service-to-service communication. It is used to decouple services, buffer spike loads (rate limiting / traffic shaping), and handle long-running backend tasks asynchronously without blocking client responses.

Question 8

Explain the difference between Monolithic and Microservices architectures.

Accepted Answer

A Monolithic architecture packages the entire application logic (UI, business layer, database access) into a single, unified codebase and deployment unit. Easy to build and test initially, but hard to scale and maintain. A Microservices architecture breaks the application into a collection of small, independent, loosely-coupled services that communicate via API protocols (REST/gRPC), enabling independent scaling and technology stacks.

Question 9

What is DNS (Domain Name System)?

Accepted Answer

DNS is the phonebook of the internet. It translates human-readable domain names (like hifitoolkit.com) into IP addresses (like 192.168.1.1) that computers use to find and route traffic to servers.

Question 10

What is API Gateway, and what are its responsibilities?

Accepted Answer

An API Gateway is a server that acts as a single entry point for all client requests in a microservices architecture. Responsibilities include: routing requests to appropriate microservices, protocol translation (e.g. HTTP to gRPC), authentication/authorization, rate limiting, and collecting metrics/logging.

Question 11

Explain Eventual Consistency.

Accepted Answer

Eventual Consistency is a consistency model used in distributed systems where updates made to a database node will replicate to all other replicas in the background. The system does not guarantee immediate consistency, but guarantees that if no new updates are made, all replicas will eventually become identical.

Question 12

What is the Database Index, and how does it work?

Accepted Answer

A database index is a data structure (commonly B-Trees or Hash Tables) that improves the speed of data retrieval operations on a database table. It acts like a book index: instead of searching every row (Full Table Scan), the database traverses the index to locate rows instantly. Indexes speed up reads but slow down writes because they must be updated on insert/update.

Question 13

Explain the difference between SQL and NoSQL databases.

Accepted Answer

SQL databases (MySQL, PostgreSQL) are relational, have predefined strict schemas, use SQL for queries, and scale vertically. Best for structured data and complex transactional integrity. NoSQL databases (MongoDB, DynamoDB, Cassandra) are non-relational, have dynamic schemas, scale horizontally, and store data in key-value, document, column, or graph structures. Best for large, unstructured, fast-changing datasets.

Question 14

What is the Circuit Breaker pattern?

Accepted Answer

The Circuit Breaker pattern prevents a cascading failure in a distributed system. If calls to a downstream service fail repeatedly, the circuit breaker 'trips' (opens), and all subsequent calls return an error instantly without wasting system threads. After a timeout, it allows a few test requests (half-open) to check if the downstream service has recovered.

Question 15

Explain rate limiting algorithms.

Accepted Answer

Rate limiting restricts the number of requests a user can make in a given timeframe. Algorithms include:
1. Token Bucket: Tokens are added to a bucket at a fixed rate. Requests consume tokens. If empty, request is blocked.
2. Leaky Bucket: Requests are queued and processed at a constant rate, smoothing out spikes.
3. Sliding Window Log: Tracks timestamps of requests, highly accurate but resource-intensive.

Question 16

What is a single point of failure (SPOF), and how do you eliminate it?

Accepted Answer

A Single Point of Failure (SPOF) is a component in a system whose failure results in the entire system crashing. It is eliminated by introducing **redundancy** at all layers (e.g. running multiple web servers behind a load balancer, setting up Multi-AZ active-passive database replication, using DNS failover).

Question 17

What is Database Replication, and what is the difference between synchronous and asynchronous replication?

Accepted Answer

Database replication copies data from a primary server to replica servers. In **Synchronous Replication**, the primary writes data and waits for all replicas to acknowledge the write before responding to the client (guarantees data consistency, but high latency). In **Asynchronous Replication**, the primary responds immediately after writing, and data is copied to replicas in the background (low latency, but risk of data loss if the primary crashes before replication).

Question 18

Explain consistent hashing.

Accepted Answer

Consistent Hashing is a distributed hashing scheme that maps both keys and servers to a circular ring. When servers are added or removed, it minimizes the number of keys that must be re-mapped (only $K/N$ keys on average, where $K$ is keys and $N$ is servers). Used in distributed caches (like Memcached) and NoSQL databases (Cassandra) to partition data efficiently.

Question 19

What is DNS Anycast?

Accepted Answer

DNS Anycast is a routing technique where multiple physical DNS servers across the globe share the same single IP address. BGP routers automatically route a client's DNS query to the physically closest server hosting that IP address, improving DNS lookup speeds and mitigating DDoS attacks.

Question 20

How do you design a URL Shortening service (like Bit.ly)?

Accepted Answer

A URL Shortening service can be designed as follows:
1. API: `POST /api/v1/shorten` returns short URL, `GET /[hash]` redirects (302 Found) to long URL.
2. Hash Generation: Base62 encode a unique auto-incrementing ID (generated by a distributed ID generator like Snowflake) to create a 7-character short hash (e.g. `62^7` supports 3.5 trillion URLs).
3. Database: Key-value NoSQL (DynamoDB/MongoDB) or Relational (PostgreSQL) mapping `hash -> long_url`.
4. Cache: Cache highly active hot redirects in Redis to minimize database read queries.

System Design Interview Master

Difficulty Filters

Tools & Options

What is the difference between Vertical Scaling and Horizontal Scaling?

What is a Load Balancer, and how does it route traffic?

Explain the CAP Theorem.

What is a Content Delivery Network (CDN), and how does it work?

Explain Database Sharding vs Partitioning.

What is the difference between ACID and BASE properties?

What is a Message Queue (MQ), and why is it used?

Explain the difference between Monolithic and Microservices architectures.

What is DNS (Domain Name System)?

What is API Gateway, and what are its responsibilities?

Explain Eventual Consistency.

What is the Database Index, and how does it work?

Explain the difference between SQL and NoSQL databases.

What is the Circuit Breaker pattern?

Explain rate limiting algorithms.

What is a single point of failure (SPOF), and how do you eliminate it?

What is Database Replication, and what is the difference between synchronous and asynchronous replication?

Explain consistent hashing.

What is DNS Anycast?

How do you design a URL Shortening service (like Bit.ly)?

Explore Related Tools

Publishing to Google Play

Machine Learning Tutorial

ASCII to Binary Converter

Binary to ASCII Converter

Binary to Octal Converter

Bulma Box Component – Simple Container with Shadow

Follow Us

Our Tools

Our Company

Special Tools