codequest / system design
Learn

The core 21 system design concepts.

Each one explained plainly, with a live simulation or an interactive diagram where it actually helps, and what to say about it in an interview. Read them in order or jump around.

  1. 01

    Horizontal scaling & load balancing

    Add identical stateless servers behind a load balancer instead of buying one bigger machine.    Live 3D
  2. 02

    Caching

    Keep hot data in memory so repeated reads skip the database, and decide what happens when it goes stale.    Live 3D
  3. 03

    Content delivery networks (CDN)

    Serve static files from servers close to the user so your own servers only handle dynamic work.    Live 3D
  4. 04

    Database replication

    Copy the database to read replicas to scale reads and survive failures, at the cost of replication lag.    Live 3D
  5. 05

    Message queues & pub/sub

    Decouple producers from consumers so slow work happens in the background and bursts are absorbed.    Live 3D
  6. 06

    Latency, throughput & p95

    Why averages lie, why latency explodes near full utilization, and how to talk about p95 and p99.    Live 3D
  7. 07

    Sharding (partitioning)

    Split one dataset across many databases by a shard key when a single primary can no longer keep up.    Diagram
  8. 08

    Consistent hashing

    Map keys to servers so that adding or removing a server moves only a small share of the keys.    Interactive
  9. 09

    CAP theorem & consistency models

    During a network partition you must choose between answering (availability) and being correct (consistency).    Interactive
  10. 10

    SQL vs NoSQL

    Relational databases, key-value, document, wide-column and graph stores: what each is good at and when to pick it.    Comparison
  11. 11

    Rate limiting

    Cap how many requests a client can make so one user or bot cannot take the system down for everyone.    Interactive
  12. 12

    Back-of-envelope estimation

    Turn users and actions into requests per second, storage and bandwidth in two minutes, the way interviewers expect.    Interactive
  13. 13

    API gateway

    One front door for every client: routing, authentication, rate limits and response shaping in one place.    Diagram
  14. 14

    Real-time: polling, SSE & WebSockets

    How servers push updates to clients, from simple polling to persistent WebSocket connections, and what each costs.    Comparison
  15. 15

    Unique ID generation

    Create IDs that never collide across many servers, sort by time, and need no central coordinator.    Interactive
  16. 16

    Idempotency

    Make retries safe, so a request sent twice (because of a timeout) has the same effect as sending it once.    Interactive
  17. 17

    Retries, timeouts & circuit breakers

    Survive slow or failing dependencies without making the outage worse.    Interactive
  18. 18

    Database indexes

    Find rows without scanning the whole table, and why every index makes writes a little slower.    Interactive
  19. 19

    Leader election & consensus

    Get several machines to agree on one leader (and one truth) even when some of them fail.    Diagram
  20. 20

    Multi-region architecture

    Run in several parts of the world for lower latency and to survive a whole region going down.    Diagram
  21. 21

    Bloom filters

    A tiny structure that answers "definitely not present" or "probably present", saving expensive lookups.    Interactive