Fundamentals for Software Architects
Distributed systems fundamentals are the theory behind data that lives on many machines. CAP and PACELC explain trade-offs between consistency, availability and latency, replication, partitioning and consensus keep copies in sync, and time, ordering and idempotency deal with the fact that clocks and networks are unreliable.
Stage: Distributed Systems · Part of the Software Architect roadmap · Last reviewed 19 November 2026
What to learn in Fundamentals
- State the CAP theorem precisely and what a partition means
- Use PACELC to reason about latency versus consistency without partitions
- Compare leader-follower and multi-leader replication
- Choose a partition key and predict hot spots
- Explain what a consensus protocol like Raft achieves
- Make an operation idempotent with an idempotency key
Fundamentals interview questions
What does the CAP theorem actually say?
During a network partition, a distributed system must choose between consistency, meaning every read sees the latest write, and availability, meaning every request gets a non-error response. Partitions cannot be avoided, so the choice is between C and A when they occur.
Why does idempotency matter in distributed systems?
Networks lose acknowledgements, so clients and brokers retry requests. An idempotent operation gives the same result when repeated, so retries do not cause duplicate payments or records. It is usually implemented with unique request keys.
Where Fundamentals fits in the Software Architect roadmap
Fundamentals belongs to the Distributed Systems stage (advanced level), which takes about 7 hours. Reason about failure.
Other units in this stage
Back to the full Software Architect roadmap