What do Byzantine generals and achieving consensus in blockchain have in common? Much more than might be assumed.

Description of the Byzantine General Problem

Blockchain networks are constructed to be as decentralised as possible. Validators are responsible for security and adding more blocks. They perform their tasks within a given consensus algorithm and are rewarded for their efforts. 

The theorem assumes the following scenario: several army units commanded by Byzantine generals have chosen to attack one selected city. The divisions are dispersed across the battlefield. Each general has to make a decision – to attack or to withdraw his troops. The decision itself is not as important as the fact that all commanders must make it in consensus to act in a way that is consistent for the entire army. The decision must be coherent, final and made in a synchronised manner. The only form of communication between them is through messengers. These, in turn, can be delayed in time or lost. It is also not excluded that some generals may, for unknown reasons, decide to act fraudulently to sabotage the assumed plan.

The role of validators in blockchain

Byzantine generals are, in this case, the counterpart of validators in the blockchain context. Their consensus on the state of the network is needed for it to work properly. Consensus is usually reached if about ⅔ of the nodes are working properly and are compatible with each other (networks have different requirements on this topic). How resistant a network is to rogue actors (e.g. nodes that want to act maliciously) is referred to as Byzantine Fault Tolerance (BFT). 

Interesting fact: a 51% attack is a type of threat in which a miner or group of miners gains control of more than 50% of the network's computing power in the case of Proof-of-Work or more than 50% of the tokens in the case of Proof-of-Stake. This allows blockchain to be manipulated by double-spending, censoring transactions, or stopping block mining. The attack threatens the integrity of the entire network. 

Solving the problem – consensus algorithms

In distributed systems, consensus algorithms are designed to solve the problem of Byzantine generals. The level of Byzantine fault tolerance depends on them. Mechanisms such as Proof-of-Work or Proof-of-Stake, for example, assume consensus about the state of the network, even though malicious actors may occur. Each mechanism has its own unique characteristics. This is an indispensable part of the blockchain as a space where no central entity controls the network. Depending on the type of algorithm, they differ in the method of reaching a consensus. However, the goal remains the same. 

Summary – blockchain trilemma vs. BFT

If you have not had a chance to read the article on the blockchain trilemma so far, you can do so here. Security, decentralisation and scalability result from an interconnected system of vessels. The coordinated operation of the validators, its speed as measured by the number of transactions per second and the level of BFT needed to update the state of the network is crucial for the network to function optimally. It is acceptable to say that the biggest challenge for projects is maintaining the highest possible scalability level. Consensus algorithms with many validators provide a very high level of security, but the challenge is to scale the network to levels that allow, for example, global micropayments. Nevertheless, the technology is constantly evolving, and in a few years, blockchain may see new chapters of innovation that are influencing the adoption and entry of crypto into the mainstream.