Paxos and Raft let nodes agree despite failures.
Consensus lets nodes agree on a value despite failures: Paxos, Raft (leader election, log replication), and Byzantine fault tolerant protocols (PBFT, blockchain). Used for configuration stores (etcd, ZooKeeper), distributed databases and coordination. Needs a majority quorum.
etcd uses Raft to keep Kubernetes cluster state consistent.
Consensus algorithms let a group of nodes agree on a value or log despite failures, which is needed for reliable replicated systems.
Paxos and Raft let nodes agree despite failures.
Consensus algorithms let a group of nodes agree on a value or log despite failures, which is needed for reliable replicated systems.
It elects a leader, which replicates a log of commands to followers, and a majority agreement commits each entry.
Two majorities always overlap, so any decision made by a majority includes at least one node that knows the previous decision.
Typical applications include electing a leader for a cluster of database nodes, keeping a replicated configuration consistent across servers and ensuring all nodes apply the same sequence of commands in a distributed store.
Study "Distributed systems" first (topic 81 of 104 in Software & Computing). It covers the groundwork this topic builds on.
Leaders and terms. Safety vs liveness. FLP impossibility.
etcd uses Raft to keep Kubernetes cluster state consistent.
The next topic is "Caching and load balancing", which builds directly on this one.
It is classed as Advanced in Software & Computing (topic 82 of 104). It is an advanced topic. It assumes you are comfortable with the intermediate topics in Software & Computing.
Want to practise? Open the interactive course for quick checks, quizzes, simulations and progress tracking.