Paper 2023/280

A Simple Single Slot Finality Protocol For Ethereum

Abstract

The implemented consensus protocol of Ethereum, Gasper, has an hybrid design: it combines a protocol that allows dynamic participation among validators, called LMD-GHOST, and a finality gadget on top, called Casper. This design has been motivated and formalized by Neu, Tas, and Tse (S&P 2021) through the introduction of the ebb-and-flow class of protocols, which are protocols with two confirmation rules that output two ledgers, one that provides liveness under dynamic participation (and synchrony), LMD-GHOST, and one that provides safety even under network partitions, Casper. Currently, Gasper takes between 64 and 95 slots to finalize blocks. Because of that, a significant portion of the chain is susceptible to reorgs. The possibility to capture MEV (Maximum Extractable Value) through such reorgs can then disincentivize honestly following the protocol, breaking the desired correspondence of honest and rational behavior. Moreover, the relatively long time to finality forces users to choose between economic security and faster transaction confirmation. This motivates the study of the so-called single slot finality protocols: consensus protocols that finalize a block in each slot and, more importantly, that finalize the block proposed at a given slot within such slot. In this work we propose a simple, non-blackbox protocol that combines a synchronous dynamically available protocol with a finality gadget, resulting in a secure ebb-and-flow protocol that can finalize one block per slot, paving the way to single slot finality within Ethereum. Importantly, the protocol we present can finalize the block proposed in a slot, within such slot.