CAP, PACELC & Consistency Models
What "consistent" really means across replicas: CAP as a partition-time choice, PACELC for the normal case, the model hierarchy from linearizable to eventual, quorums, anti-entropy, and conflict resolution with CRDTs.
About 85 minutes · 16 exam questions · 9 interview questions · 2 code labs
What you will learn
- The problem: many copies, one truth?
- CAP, properly understood
- PACELC: the trade-off you pay even without partitions
- The consistency model ladder: linearizable → causal → eventual
- Linearizability vs serializability (and strict serializability)
- Quorums: N, R, W, sloppy quorums and hinted handoff
- Anti-entropy: read repair, Merkle trees, and why repairs are mandatory
- Conflict resolution: LWW, version vectors and CRDTs
- Choosing consistency per feature (the interview move that scores)