Previous Module
Tendermint Consensus

🛡️ BFT Fault Tolerance: The Byzantine Generals Problem

Learn how blockchains reach consensus despite malicious actors

🛡️ What is Byzantine Fault Tolerance?

Byzantine Fault Tolerance (BFT) is the ability of a distributed system to reach consensus even when some nodes are faulty, malicious, or sending contradictory information. It's named after the Byzantine Generals Problem - a classic computer science challenge.

Why BFT Matters in Blockchain

🌐

Trustless Networks

Blockchain operates in adversarial environments where nodes cannot trust each other. BFT ensures consensus despite malicious actors.

🔒

Security Guarantee

BFT algorithms mathematically prove that consensus is safe as long as fewer than 1/3 of nodes are Byzantine (malicious).

Instant Finality

Unlike PoW which has probabilistic finality, BFT consensus provides instant, deterministic finality. No waiting for confirmations.

🔍 Interactive: The Byzantine Challenge

Explore the core challenges that Byzantine Fault Tolerance solves:

⚔️

The Challenge

Byzantine generals must coordinate an attack, but communication is difficult and trust is uncertain.

The Challenge:

How do distributed parties reach agreement when they cannot meet in person?

Types of Faults

💥

Crash Faults

Nodes that fail by stopping (crash, network disconnect). They don't send any messages after failure.

Example: Server crashes, loses power, or goes offline
🎭

Byzantine Faults

Nodes that behave arbitrarily or maliciously. They can send conflicting messages, lie, or collude.

Example: Hacked node, malicious validator, software bug
💡

The 1/3 Threshold

BFT consensus requires more than 2/3 of nodes to be honest. This means the system can tolerate up to 1/3 Byzantine (faulty/malicious) nodes while still reaching correct consensus.

Byzantine nodes: f

Total nodes: n ≥ 3f + 1

Example: 7 nodes can tolerate 2 Byzantine nodes

BFT vs Other Consensus

FeatureBFTPoWPoS
Fault Tolerance< 1/3 Byzantine< 51% hashrate< 51% stake
FinalityInstantProbabilisticNear-instant
Energy UseVery LowVery HighLow
ThroughputHigh (1,000+ TPS)Low (~7 TPS)Medium (~30 TPS)