Distributed Systems

Consistency, Availability, Partition Tolerance — the fundamental tradeoff in distributed systems

Extends CAP: during partitions choose A or C; else choose Latency or Consistency

Linearizability, sequential, causal, eventual consistency, and session guarantees

Safe retries, idempotency keys, delivery guarantees, Kafka exactly-once, deduplication patterns

Redis SETNX, Redlock, fencing tokens, ZooKeeper znodes, and when not to use locks

Bully algorithm, Raft election, ZooKeeper znodes, split-brain prevention, STONITH

Fan-out convergence, phi accrual failure detection, SWIM, Merkle tree anti-entropy

R + W > N overlap rule, tuning configurations, sloppy quorum, hinted handoff, read repair

Orchestration vs choreography, compensating transactions, execution log, and isolation countermeasures

Dual-write problem, CDC with Debezium, at-least-once delivery, and inbox deduplication

Lamport timestamps, vector clocks, version vectors, and Hybrid Logical Clocks

Active-passive, active-active, geo-routing, CRDTs, conflict resolution, and data sovereignty