Table of Contents
Fetching ...

SoK: DAG-based Consensus Protocols

Mayank Raikwar, Nikita Polyanskii, Sebastian Müller

TL;DR

The paper addresses the scalability and performance limitations of traditional blockchains by systematizing DAG-based consensus protocols within a CAP-inspired framework. It classifies protocols into availability-focused and consistency-focused families, surveys representative schemes (structured/unstructured DAGs and optimistic/certified DAGs), and analyzes attack vectors, desirable properties, and evolving design trade-offs. Key contributions include a comprehensive taxonomy, security considerations, and a forward-looking agenda highlighting performance evaluation, fairness, and privacy challenges in high-throughput DAG-based DLTs. The work provides a structured reference for researchers and practitioners to compare DAG protocols, identify gaps, and guide future developments with practical relevance for scalable, fair, and robust distributed ledgers.

Abstract

This paper is a Systematization of Knowledge (SoK) on Directed Acyclic Graph (DAG)-based consensus protocols, analyzing their performance and trade-offs within the framework of consistency, availability, and partition tolerance inspired by the CAP theorem. We classify DAG-based consensus protocols into availability-focused and consistency-focused categories, exploring their design principles, core functionalities, and associated trade-offs. Furthermore, we examine key properties, attack vectors, and recent developments, providing insights into security, scalability, and fairness challenges. Finally, we identify research gaps and outline directions for advancing DAG-based consensus mechanisms.

SoK: DAG-based Consensus Protocols

TL;DR

The paper addresses the scalability and performance limitations of traditional blockchains by systematizing DAG-based consensus protocols within a CAP-inspired framework. It classifies protocols into availability-focused and consistency-focused families, surveys representative schemes (structured/unstructured DAGs and optimistic/certified DAGs), and analyzes attack vectors, desirable properties, and evolving design trade-offs. Key contributions include a comprehensive taxonomy, security considerations, and a forward-looking agenda highlighting performance evaluation, fairness, and privacy challenges in high-throughput DAG-based DLTs. The work provides a structured reference for researchers and practitioners to compare DAG protocols, identify gaps, and guide future developments with practical relevance for scalable, fair, and robust distributed ledgers.

Abstract

This paper is a Systematization of Knowledge (SoK) on Directed Acyclic Graph (DAG)-based consensus protocols, analyzing their performance and trade-offs within the framework of consistency, availability, and partition tolerance inspired by the CAP theorem. We classify DAG-based consensus protocols into availability-focused and consistency-focused categories, exploring their design principles, core functionalities, and associated trade-offs. Furthermore, we examine key properties, attack vectors, and recent developments, providing insights into security, scalability, and fairness challenges. Finally, we identify research gaps and outline directions for advancing DAG-based consensus mechanisms.

Paper Structure

This paper contains 45 sections, 2 figures, 3 tables.

Figures (2)

  • Figure 1: Blockchain and blockDAG
  • Figure 2: The evolution of DAG-based consensus protocols over time. Our estimation of the total number of protocols is based on a manual review of papers citing previous work starting with Hashgraph baird2020hashgraph and Tangle popov2018tangle—in Google Scholar and their related work sections. This count, intended as a rough approximation, highlights the growing diversity and interest in DAG protocols.