Learning Objectives – after this chapter you will be able to:
- Explain what consensus means and why distributed ledgers need it.
- Describe the Byzantine Generals Problem and Byzantine fault tolerance.
- Distinguish safety, liveness and the two kinds of finality.
- Compare common consensus mechanisms: PoW, PoS, DPoS, PoA and PBFT.
Consensus Mechanism
- Consensus means that every honest participant of the network ends up with the same information – the same ordered ledger.
- A consensus mechanism is the method a distributed network uses to validate new entries, decide their order and keep the shared database secure, even when some participants are faulty or dishonest.
- For a cryptocurrency, the shared database is the blockchain – so the consensus mechanism is what secures the blockchain.
- Proof-of-Work (PoW) and Proof-of-Stake (PoS) are the two best-known consensus mechanisms for processing transactions and creating new blocks.
- Both help blockchains synchronise data, validate information and process transactions. Each has proven able to maintain a large blockchain, each has pros and cons, and they take very different approaches (Chapters 15–17).
The Byzantine Generals Problem
In 1982 Lamport, Shostak and Pease described generals of the Byzantine army surrounding a city. They can communicate only by messengers and must agree on a common plan (attack or retreat), but some generals may be traitors who send conflicting messages. How can the loyal generals still agree?
- In a blockchain the “generals” are nodes, the “messages” are blocks and votes, and the “traitors” are faulty or malicious nodes.
- A system that keeps working correctly despite such nodes is Byzantine Fault Tolerant (BFT). Classical BFT protocols tolerate up to f faulty nodes out of 3f + 1 total – i.e. fewer than one-third.
- Open (public) blockchains also face Sybil attacks – one person creating many fake identities. PoW and PoS solve this by tying influence to a scarce resource (computing power or staked coins) rather than to the number of identities.
Properties of Consensus
| Safety (consistency) | Honest nodes never finalize conflicting blocks – “nothing bad happens”. |
| Liveness | The network keeps adding new blocks – “something good eventually happens”. |
| Probabilistic finality | A block becomes more secure as more blocks are built on top of it (Bitcoin PoW). Reversal is never strictly impossible, just increasingly unlikely. |
| Deterministic (economic) finality | Once finalized, a block cannot be reverted unless a large share of validators break the rules and are penalised (PBFT-style protocols, Ethereum PoS). |
Common Consensus Mechanisms
| Mechanism | How the block producer is chosen | Typical use |
|---|---|---|
| Proof-of-Work (PoW) | Miners compete to solve a hash puzzle; the first valid solution wins. | Bitcoin, Litecoin |
| Proof-of-Stake (PoS) | Validators lock up coins; proposers are chosen pseudo-randomly, weighted by stake. | Ethereum, Cardano |
| Delegated PoS (DPoS) | Token holders vote for a small set of delegates who produce blocks. | EOS, TRON |
| Proof-of-Authority (PoA) | A fixed set of approved, identified validators take turns. | Private / consortium networks, some testnets |
| PBFT (Practical BFT) | Known validators exchange several rounds of votes; a block is final once 2/3+ agree. | Permissioned ledgers (e.g. Hyperledger family) |
| Proof of History (PoH) | Not a full consensus on its own: a verifiable clock that orders events, used together with PoS. | Solana |
Every consensus design balances security, decentralization and performance. Permissioned networks can use fast voting protocols because validators are known; public networks need Sybil-resistant mechanisms such as PoW or PoS.
Key Takeaways
- Consensus lets untrusting nodes agree on one valid, ordered ledger.
- BFT systems tolerate fewer than one-third faulty nodes; PoW/PoS add Sybil resistance for open networks.
- Finality can be probabilistic (PoW) or deterministic/economic (PBFT, Ethereum PoS).
Question / Answer (15 questions)
Answer : A consensus mechanism is the set of rules and processes by which the nodes of a distributed network validate new entries, agree on their order and keep a single, consistent ledger – even if some nodes are faulty or malicious.
Answer : There is no central authority to decide which transactions are valid or which block comes next. Without consensus, nodes could hold different versions of the ledger and the same coins could be spent twice.
Answer : It is a thought experiment in which several generals must agree on a common plan by sending messages, while some generals are traitors who send conflicting messages. It models how distributed nodes can reach agreement when some participants are unreliable or dishonest.
Answer : BFT is the ability of a distributed system to keep operating correctly even when some nodes fail or act maliciously. Classical BFT protocols can tolerate up to f faulty nodes among 3f + 1 nodes, i.e. fewer than one-third.
Answer : In a Sybil attack one entity creates many fake identities to gain influence. PoW and PoS resist it by tying influence to a scarce resource – computing power or staked coins – so creating many identities does not increase power.
Answer : Safety means honest nodes never finalize conflicting versions of the ledger. Liveness means the network continues to make progress by adding new blocks. A good consensus protocol must provide both.
Answer : In probabilistic finality, used by Bitcoin PoW, a block becomes harder to reverse as more blocks are added on top. Reversal becomes increasingly unlikely but is never mathematically impossible.
Answer : Once a block is finalized it cannot be reverted unless a large fraction of validators violate the protocol. In Ethereum PoS, doing so would cause a large amount of their staked ETH to be slashed.
Answer : Token holders vote to elect a limited number of delegates (block producers). These delegates take turns producing blocks. DPoS is fast but more centralized because only a few producers exist.
Answer : In PoA, a fixed set of known and approved validators produce blocks, staking their identity and reputation. It is efficient and is used in private or consortium networks and some test networks.
Answer : Practical Byzantine Fault Tolerance (Castro and Liskov, 1999) lets a known set of nodes agree through several rounds of message exchange. A decision is committed once more than two-thirds of nodes agree, providing immediate finality, but message overhead limits the number of nodes.
Answer : No. Proof of History is a cryptographic clock that creates a verifiable order of events. Solana uses it together with a Proof-of-Stake-based consensus.
Answer : In permissioned networks, validators are known and limited, so voting protocols are practical and Sybil attacks are not a concern. Public networks allow anyone to join, so they need Sybil-resistant mechanisms based on scarce resources.
Answer : Security (resistance to attacks), decentralization (how many independent parties take part) and performance (throughput and latency). Improving one often affects the others.
Answer : Both are used to choose who creates the next block, both synchronise data and validate transactions across a public network, both provide Sybil resistance through a scarce resource, and both reward honest participants.
MCQ Test : Check Your Understanding (5 questions)
Select an option and press Check Answer, or answer all questions and press Submit Test to see your score. Each answer comes with an explanation.
Bibliography/References
The notes, questions and MCQs on this page are original explanations written for this course. The works listed below were consulted and are cited for attribution and further reading. Product names, logos and trademarks belong to their respective owners. Short quotations, where used, are marked and attributed.
- Lamport, L., Shostak, R., & Pease, M. (1982). The Byzantine Generals Problem. ACM Transactions on Programming Languages and Systems, 4(3), 382–401. https://doi.org/10.1145/357172.357176
- Castro, M., & Liskov, B. (1999). Practical Byzantine Fault Tolerance. Proceedings of the Third Symposium on Operating Systems Design and Implementation (OSDI ’99). https://pmg.csail.mit.edu/papers/osdi99.pdf
- Nakamoto, S. (2008). Bitcoin: A Peer-to-Peer Electronic Cash System. https://bitcoin.org/bitcoin.pdf
- ethereum.org. (n.d.). Consensus mechanisms. Ethereum Developer Documentation. https://ethereum.org/en/developers/docs/consensus-mechanisms/
- Yaga, D., Mell, P., Roby, N., & Scarfone, K. (2018). Blockchain Technology Overview (NISTIR 8202). National Institute of Standards and Technology. https://doi.org/10.6028/NIST.IR.8202
- Investopedia. (n.d.). What Is Proof of Work (PoW) in Blockchain? https://www.investopedia.com/terms/p/proof-work.asp
- Investopedia. (n.d.). What Does Proof-of-Stake (PoS) Mean in Crypto? https://www.investopedia.com/terms/p/proof-stake-pos.asp
- Narayanan, A., Bonneau, J., Felten, E., Miller, A., & Goldfeder, S. (2016). Bitcoin and Cryptocurrency Technologies: A Comprehensive Introduction. Princeton University Press. ISBN 978-0-691-17169-2.