Byzantine Fault Tolerance

The name itself traces back to a classic thought experiment in computer science: the Byzantine Generals Problem. The classic version sees a group of generals laying siege to an enemy city, able to communicate with each other only via messenger. Together they try to agree on a plan: attack or retreat. However, some of those generals may be traitors attempting to sabotage the plan by sending additional, conflicting messages to different parties.

Byzantine fault tolerance is the mathematical assurance that, as long as fewer than one-third of the generals are traitors, two conflicting plans cannot both receive approval from more than two-thirds of the group. This is because any two groups large enough to approve a plan must share at least one honest general, who would never approve contradictory orders. The consensus system will therefore not approve conflicting outcomes.

When applied to blockchain tech, this same logic ensures that decentralized ledgers can remain secure and coherent without a central authority to manage them. The generals are replaced with network nodes; any single compromised node cannot force the network to finalize conflicting versions of the ledger. This is how many independent network participants are able to agree on which transactions are legitimate, even if some bad actors are among them.

Instead of relying on the honesty of every participant, BFT-based blockchains are therefore able to cryptographically and mathematically ensure that a two-thirds majority of honest participants can never be overruled by a minority of bad actors, thereby maintaining the integrity of the network. These guarantees make BFT-based consensus particularly valuable for financial applications, supply-chain management and other real-world applications.