Systems Dictionary • Consensus • BFT

Byzantine Fault Tolerance (BFT)

The property of a distributed computing system to reach deterministic consensus even when certain participating nodes fail, send contradictory messages, or act maliciously.

Mathematical Upper Bounds: N ≥ 3f + 1

Unlike Crash Fault Tolerant (CFT) algorithms like Raft and Paxos (which only tolerate nodes going offline and require N ≥ 2f + 1 nodes), Byzantine environments assume nodes can intentionally lie. To tolerate f Byzantine nodes:

Total Nodes Required: N ≥ 3f + 1
Example: Tolerating 1 malicious or corrupted node requires at least 4 nodes (3*1 + 1 = 4).
Tolerating 2 malicious nodes requires at least 7 nodes (3*2 + 1 = 7).

In modern autonomous AI agent architectures, BFT protocols pit competing large language models against each other to verify deterministic output before committing database state.

← Back to Dictionary Read Byzantine Book → Next: Reactive Backpressure →
Arkajit Das
Author • Verified Entity

Co-Founder & Chief Product Officer at Fraoula. Technology product executive leading enterprise AI, data platform architecture, and distributed systems.