Lawrence Jenger
February 27, 2025 07:33
A comprehensive analysis of blockchain resilience to adversarial control, examining the limits of safety and activity of various client and network models.
In the evolving situation of blockchain technology, the resilience of blockchain systems to adversarial control remains crucial. Recent explorations by A16Z Crypto delve into the key issue of the number of malicious validators that blockchains can withstand, while maintaining their core properties of vibrancy and security. This study examines whether the threshold is 50%, 33%, or 99%, as suggested by Ethereum co-founder Vitalik Buterin.
Exploring the resilience of blockchain
This study considers examining dimensions such as whether clients and valiters are always active or sometimes dormant, as well as the nature of network synchronization, highlighting the importance of modeling blockchain clients. This study, conducted in collaboration with Dionysis Zindros and David TSE, systematically classifies consensus models across these dimensions to depict achievable safety and activity resilience.
State Machine Replication and Consensus Protocol
The State Machine Replication (SMR) consensus protocol is essential for blockchain functionality and ensures that transactions are executed in a consistent order across the network. These protocols must be Byzantine fault resistance (BFT). This study investigates the largest acceptable adversarial percentage, and explores both classical and partially synchronized networks.
Determining acceptable validator controls
This study questions the upper limit of safety and resilience that any blockchain protocol can achieve. It emphasizes that the network’s ability to ensure that messages are delivered between validators affects these limitations. Synchronized networks allow for greater resilience compared to partially synchronous networks where network latency can disrupt communications.
Client modeling and network synchronization
Client characteristics such as communication capabilities and activity levels have a significant impact on resilience thresholds. This study proposes a model of BFT SMR consensus based on the characteristics of these clients and networks, providing a comprehensive analysis of resilience achievable in a variety of scenarios.
To dig deeper into the findings, including detailed models and theoretical proofs, the complete paper is available in the IACR EPRINT archive. This research not only integrates existing knowledge, but also introduces new protocols and impossible theorems, which contributes greatly to understanding blockchain security.
For more information, see the original article on the A16Z Crypto.
Image source: ShutterStock
