Jincheng Zhang
Distributed consensus algorithms are fundamental to many critical systems, including blockchain networks, sensor networks, and distributed databases. However, these systems are vulnerable to Byzantine faults, where malicious nodes can arbitrarily deviate from the agreed-upon protocol. Verifying the convergence and correctness of consensus algorithms under these conditions is a notoriously difficult problem. This paper presents a novel approach to probabilistic formal verification of distributed consensus algorithms with Byzantine fault tolerance. We model the consensus algorithm as a stochastic process and leverage probability covers and Markov chain analysis to derive rigorous proofs of convergence and fault tolerance. This method allows us to quantify the probability of correct operation even in the presence of arbitrary malicious behavior, offering a significant advancement over traditional approaches that often rely on idealized assumptions. The key contribution lies in the ability to provide probabilistic guarantees for consensus algorithm behavior, rather than simply demonstrating eventual convergence. We illustrate the application of this framework with a simplified example, highlighting its potential for scaling to more complex consensus protocols.