FairDAG: Consensus Fairness over Multi-Proposer Causal Design

TL;DR

FairDAG addresses adversarial transaction-order manipulation in blockchains by layering fairness atop DAG-based consensus, enabling multi-proposer blocks and avoiding single-leader bottlenecks. It introduces two fairness variants: FairDAG-AB, which achieves Ordering Linearizability via an absolute final ordering, and FairDAG-RL, which achieves $\gamma$-Batch-Order-Fairness through a relative dependency-graph approach. Key innovations include the DAG Layer (multi-proposer dissemination/commit with strong and weak edges) and the Fairness Layer (AOI-based ordering for AB and a Hamiltonian path over a dependency graph for RL), plus an Ordering Indicator Manager to handle uncommitted vertices and enforce fairness under Byzantine faults. Analytical and empirical evaluations show that FairDAG variants outperform Pompe and Themis in throughput and fairness quality, validating resilience against adversarial ordering while maintaining high performance in a DAG-based setting.

Abstract

The rise of cryptocurrencies like Bitcoin and Ethereum has driven interest in blockchain database technology, with smart contracts enabling the growth of decentralized finance (DeFi). However, research has shown that adversaries exploit transaction ordering to extract profits through attacks like front-running, sandwich attacks, and liquidation manipulation. This issue affects blockchains where block proposers have full control over transaction ordering. To address this, a more fair transaction ordering mechanism is essential. Existing fairness protocols, such as Pompe and Themis, operate on leader-based consensus protocols, which not only suffer from low throughput caused by the single-leader bottleneck, but also allow adversarial block proposers to manipulate transaction ordering. To address these limitations, we propose a new framework, FairDAG, that runs fairness protocols on top of DAG-based consensus protocols. FairDAG improves protocol performance in both throughput and fairness quality by leveraging the multi-proposer design and validity property of DAG-based consensus protocols. We conducted a comprehensive analytical and experimental evaluation of two FairDAG variants - FairDAG-AB and FairDAG-RL. Our results demonstrate that FairDAG outperforms prior fairness protocols in both throughput and fairness quality.

Figures (5)

• Figure 1: Condorcet Cycle
• Figure 2: $T_1$ will be ordered before $T_4$ if Ordering Linearizability holds regardless of the local ordering from $R_4$.
• Figure 3: A final ordering of six transactions that satisfies $\gamma$-Batch-Order-Fairness with $\gamma = \frac{2}{3}$, 3 correct replicas, and 1 Byzantine replica.
• Figure 4: Network topology with Clients, Replicas, and Networks. $Rsp_{i}$ represents a client response for transaction $T_i$, including the execution results of $T_i$.
• Figure 5: Architecture of FairDAG: (a) Replicas reliably broadcast blocks containing their local ordering fragments. (b) Each replica receives blocks delivered through Reliable Broadcast. (c) Each replica forms a local view of the DAG using received blocks and reference links, where different colors represent the subdags of different committed leader vertices. (d) Local orderings in subdag $A_r$ are used as input of the fairness layer after $L_r$ is committed. (e) Finalize transaction ordering using absolute ordering mechanism (left) or relative ordering mechanism (right), based on the committed local orderings. (f) A final transaction ordering is generated.

Theorems & Definitions (4)

• definition 1
• definition 2
• definition 3
• definition 4