The building blocks of system design interviews and real distributed systems: scaling, load balancing, caching, data partitioning, consistency models and event-driven architecture. Every concept is presented as a trade-off, because that is how it is actually assessed.
You’re reading a free preview. Sign up to see every topic and question — and to play this deck as games.
Vertical and horizontal scaling, what statelessness buys you, and the estimation habits that make the rest of system design tractable.
38 total questions
Card #1
Vertical scaling (scaling up)
Card #2
Why statelessness makes a service easy to scale
Card #3
Sticky sessions (session affinity)
1. A service is described as 'stateless'. What follows from that?
2. The main limitation of vertical scaling is that:
3. A service reports a p99 latency of 800 ms. That means:
1. Adding CPU, RAM or disk to a single existing machine is called ___ scaling.
Hint: The opposite of horizontal
2. A service that keeps no request-specific data in an instance's memory is described as ___.
Hint: Any replica can serve any request
3. The latency value that only 1% of requests exceed is the ___ latency.
Hint: A percentile
