# Bullshark: DAG BFT Protocols Made Practical Content type: Research Paper Summary: Bullshark introduces an asynchronous Byzantine Fault Tolerant (BFT) consensus protocol built upon a Directed Acyclic Graph (DAG) architecture. The authors address the latency and throughput limitations of traditional linear-blockchain consensus mechanisms by decoupling the data dissemination layer from the consensus ordering layer. By utilizing a DAG to structure the propagation of transactions, Bullshark allows validators to reach consensus on the order of transactions asynchronously. The paper Key concepts: Alexander Spiegelman, Neil Giridharan, Alberto Sonnino, Lefteris Kokoris-Kogias, dag, bft, consensus, asynchronous, scalability, distributed systems
Paperadvanced

Bullshark: DAG BFT Protocols Made Practical

Alexander Spiegelman, Neil Giridharan, Alberto Sonnino, Lefteris Kokoris-Kogias

arXiv preprintJanuary 2022

Abstract

Bullshark introduces an asynchronous Byzantine Fault Tolerant (BFT) consensus protocol built upon a Directed Acyclic Graph (DAG) architecture. The authors address the latency and throughput limitations of traditional linear-blockchain consensus mechanisms by decoupling the data dissemination layer from the consensus ordering layer. By utilizing a DAG to structure the propagation of transactions, Bullshark allows validators to reach consensus on the order of transactions asynchronously. The paper demonstrates that this approach significantly enhances throughput while maintaining safety and liveness under asynchronous network conditions. It presents a modular architecture where the protocol can achieve high performance by delegating the ordering of blocks to a specific DAG-based structure, which avoids the bottleneck of sequential block production. The contribution is pivotal for scalable decentralized systems aiming for sub-second latency in highly parallelized environments, proving that high-throughput BFT is practical at scale.

Key Findings

  • 1Decouples data dissemination from consensus ordering to improve scalability.
  • 2Utilizes a DAG structure to achieve asynchronous BFT performance.
  • 3Maintains high throughput in adverse network conditions.
  • 4Demonstrates sub-second latency in practical deployment scenarios.

Topics

dagbftconsensusasynchronousscalabilitydistributed systems
Download / View

Citation

Alexander Spiegelman, Neil Giridharan, Alberto Sonnino, Lefteris Kokoris-Kogias (2022). Bullshark: DAG BFT Protocols Made Practical. arXiv preprint. https://arxiv.org/abs/2209.05633

BibTeX

@misc{bullsharkdag2022,
  title = {Bullshark: DAG BFT Protocols Made Practical},
  author = {Alexander Spiegelman and Neil Giridharan and Alberto Sonnino and Lefteris Kokoris-Kogias},
  year = {2022},
  howpublished = {\url{https://arxiv.org/abs/2209.05633}},
}

Knowledge Explorer

Explore Related Concepts

See how Bullshark: DAG BFT Protocols Made Practical connects to glossary terms, books, and other research in the Knowledge Graph.

Explore Connections

Related Glossary Terms

OracleQuorumProposal