Angelfish Consensus: Optimizing Throughput and Latency in BFT Protocols → Research

A high-tech, white modular apparatus is depicted in a state of connection, with two primary sections slightly apart, showcasing complex internal mechanisms illuminated by intense blue light. A brilliant, pulsating blue energy stream, representing a secure data channel, actively links the two modules

A detailed close-up shot showcases a sleek, metallic apparatus immersed in a vibrant blue, viscous fluid, with white foam actively forming around its components. The image highlights the precision engineering of the device, featuring polished surfaces and intricate mechanical connections

Briefing

Modern blockchain systems leveraging Byzantine Fault Tolerant (BFT) consensus protocols have long contended with a fundamental trade-off → protocols optimized for minimal latency typically rely on a single leader for both data dissemination and consensus, while those prioritizing maximal throughput often employ a separate, asynchronous data dissemination layer with Directed Acyclic Graphs (DAGs). This research introduces Angelfish, a groundbreaking hybrid consensus protocol that seamlessly navigates this design spectrum, dynamically adjusting its operational mode to deliver both state-of-the-art peak throughput and the low latency characteristic of leader-based systems under moderate loads. The profound implication of this new theory is the potential for blockchain architectures to overcome a critical performance bottleneck, enabling more responsive and scalable decentralized applications without compromising on transactional capacity.

The image prominently displays multiple blue-toned, metallic hardware modules, possibly server racks or specialized computing units, arranged in a linear sequence. A striking blue, translucent, gel-like substance flows dynamically between these components, while white, fibrous material adheres to their surfaces

Context

Prior to this research, the prevailing theoretical limitation in eventually-synchronous BFT consensus protocols was the inherent performance dichotomy. Protocols like HotStuff excelled in low latency due to their leader-driven structure, yet often sacrificed peak throughput. Conversely, DAG-based protocols, such as Bullshark and Sailfish, achieved high throughput by decoupling data dissemination, but typically introduced higher latency.

This created a persistent challenge for blockchain architects seeking a single consensus mechanism capable of delivering both optimal responsiveness and high transaction volume across varying network conditions and loads. The field lacked a unified approach that could dynamically adapt to provide the “best of both worlds” in terms of performance.

The image presents a meticulously rendered abstract mechanism, featuring polished silver cylindrical components, a prominent blue multi-bladed rotor, and clear, transparent conduits that intricately wrap around the central elements. These components are dynamically arranged against a smooth, gradient dark grey background, highlighting their interconnectedness

Analysis

Angelfish’s core mechanism lies in its adaptive hybrid design, which fundamentally differs from previous fixed-paradigm approaches. The protocol dynamically selects a subset of participating nodes to issue lightweight votes via best-effort broadcast, rather than mandating the reliable, and thus costlier, broadcast of full DAG vertices. This novel primitive allows Angelfish to fluidly transition between a leader-based mode, which minimizes communication overhead and latency under lighter network loads, and a DAG-based mode, which maximizes parallel data dissemination and throughput during periods of high activity.

By intelligently adjusting its communication strategy, Angelfish reduces network congestion, helps lagging nodes synchronize more efficiently, and ensures consistent performance across a wide range of operational demands. The protocol’s logic is rooted in optimizing communication efficiency and adaptability, rather than adhering to a single, rigid consensus structure.

A sophisticated, futuristic circular device with luminous blue elements and intricate metallic structures dominates the frame. A vibrant cloud of white mist, interspersed with brilliant blue granular particles, actively emanates from its central core, suggesting an advanced operational process

Parameters

Core Concept → Hybrid BFT Consensus
System/Protocol Name → Angelfish
Key Authors → Qianyu Yu, Giuliano Losa, Nibesh Shrestha, Xuechao Wang
Key Performance Metrics → Optimal Throughput, Low Latency
Underlying Mechanisms → Leader-based consensus, DAG-based consensus, Dynamic voting

The image displays a close-up of metallic structures integrated with translucent blue fluid channels. The composition highlights advanced engineering and material science

Outlook

The introduction of Angelfish marks a significant step forward for BFT consensus research, opening new avenues for dynamically adaptive protocols. Future work will likely explore more sophisticated adaptive algorithms, potentially incorporating machine learning to predict network conditions and optimize protocol transitions. In 3-5 years, this theory could unlock real-world applications requiring both high transaction speeds and immediate finality, such as high-frequency decentralized finance (DeFi) exchanges, real-time supply chain tracking, and highly responsive gaming platforms on blockchain. This research also encourages further investigation into hybrid models that combine different cryptographic primitives or network topologies to achieve unprecedented performance benchmarks in decentralized systems.

Angelfish fundamentally advances blockchain consensus by demonstrating that optimal throughput and low latency are simultaneously achievable through dynamic protocol adaptation.

Signal Acquired from → Cool Papers