Modeling the Fork Probability of Blockchains: Did EIP-1559 Improve Ethereum?

Abstract

There is a natural tendency for many public blockchain systems to undergo forks. When they are resolved, they result in a chain reorganization (reorg) where the transactions from the side fork are discarded. This causes delays since clients are required to wait for more confirmations before processing transactions. In order to improve security, speed, and efficiency of any blockchain network, it is desirable to reduce the probability of forking as much as possible. For this purpose, it is necessary to have a thorough understanding of the behavior of blockchain systems in order to optimize parameters in accordance with this goal [1], [2]. This study proposes a new fork model that incorporates parameters not previously considered, such as network delay and validation degree. We have conducted several experiments to verify the validity of our proposed method, using the BlockSim simulator on the Ethereum network and EIP-1559. According to the results of this study, decreasing the validation degree and increasing the miners' marginal cost (as in EIP-1559) reduces the likelihood of a fork occurring by around 10 percent. Additionally, the experimental results validate the precision of our method in predicting the probability of forking.