Fides: Scalable Censorship-Resistant DAG Consensus via Trusted Components

TL;DR

Fides tackles scalability and censorship-resilience in asynchronous DAG-based BFT by integrating four TEEs-assisted components (MIC, RAC, RANG, TRAD) into a DAG framework. By reducing the replication factor to $n=2f+1$ and removing heavy threshold-cryptography, it achieves near-linear message complexity and robust censorship-resistance. Empirical results on ResilientDB demonstrate high throughput in both geo-distributed ($ ext{TPS} rightarrow 400{,}000$) and local networks ($ ext{TPS} rightarrow 810{,}000$), outperforming established protocols like Tusk, RCC, HotStuff, and PBFT. The design emphasizes a modular Trusted Computing Base (TCB) by embedding only essential components in TEEs, making it practical for real-world deployment in distributed blockchain systems.

Abstract

Recently, consensus protocols based on Directed Acyclic Graph (DAG) have gained significant attention due to their potential to build robust blockchain systems, particularly in asynchronous networks. In this paper, we propose Fides, an asynchronous DAG-based BFT consensus protocol that leverages Trusted Execution Environments (TEEs) to tackle three major scalability and security challenges faced by existing protocols: (i) the need for a larger quorum size (i.e., at least 3x larger) to tolerate Byzantine replicas, (ii) high communication costs and reliance on expensive cryptographic primitives (i.e., global common coin) to reach agreement in asynchronous networks, and (iii) poor censorship resilience undermining the liveness guarantee. Specifically, Fides adopts four trusted components-Reliable Broadcast, Vertex Validation, Common Coin, and Transaction Disclosure-within TEEs. Incorporating these components enables Fides to achieve linear message complexity, guaranteed censorship resilience, 2x larger quorum size, and lightweight common coin usage. Besides, abstracting these essential components rather than porting the entire protocol into TEE can significantly reduce the Trusted Computing Base (TCB). Experimental evaluations of Fides in local and geo-distributed networks demonstrate its superior performance compared to established state-of-the-art protocols such as Tusk, RCC, HotStuff, and PBFT. The results indicate that Fides achieves a throughput of 400k transactions per second in a geo-distributed network and 810k transactions per second in a local network. Our analysis further explores the protocol's overhead, highlighting its suitability and effectiveness for practical deployment in real-world blockchain systems.

Figures (2)

• Figure 1: Example of commit rule in Fides. If the leader has less than $f+1$ references from its third round (red in the figure), they are ignored; otherwise (blue in the figure), the algorithm searches the causal DAG to commit all preceding leaders (dark yellow and red in the figure) and totally orders the rest of the DAG by the predefined rule afterward.
• Figure 2: Overview Architecture of Fides