Consensus
In plain terms
Getting distributed nodes to agree on a single value or state.
Going deeper
Consensus algorithms let a cluster agree despite failures and network issues, ensuring consistency for things like leader election and committed data. It's a core, hard problem in distributed systems.
Under the hood
Must tolerate crashes (and sometimes Byzantine faults); FLP impossibility shows no async algorithm guarantees both safety and liveness under failures.
Examples
Raft, Paxos, leader election, etcd, ZooKeeper, blockchain consensus.
Think of it like...
Like a jury that must reach a unanimous verdict even when some members step out of the room temporarily.