LBFT-DAG: A Swift, Leader-Driven, DAG-Based Consortium Blockchain with Byzantine Fault-Tolerance
Xuewen Dong, Yi Liu, Teng Li, Xiaojie Guo, Youliang Tian, Yulong Shen, Xiaojiang Du
Abstract
Due to their parallel processing capabilities across different nodes, Directed Acyclic Graph (DAG) based blockchains have emerged as promising alternatives to traditional blockchains with a single-chain structure. However, existing DAG ledgers mostly have complex structures and rely on resource-intensive consensus protocols (e.g., PoW) to establish a stable main chain and determine the total order of all blocks. Moreover, they require a certain number of voting or successive PoW-like confirmations for each new block, resulting in significant delays. To tackle above issues, this paper proposes LBFT-DAG, the first Leader-drtven, Byzantine Fault- Iolerant, DAG-based consortium blockchain. In LBFT-DAG, a concise ledger consists of the leader node's main subchain and several normal nodes' parallel subchains, where each block only refers to the preceding blocks in two subchains. To quickly establish the total order of the ledger, we present a two-stage ordering algorithm following the principle of the main subchain first. Additionally, an asynchronous, Byzantine Fault-tolerant voting process on only leader blocks is designed to smoothly achieve security for all blocks. Extensive analysis confirms the security of LBFT-DAG, and experimental results demonstrate that LBFT-DAG achieves about four times higher transaction throughput on average compared to state-of-the-art solutions.
Ask about this paper
Ask your agent about it.
Lune has read the top-tier papers around this one, so every answer names the papers it rests on.
Your agent calls
Lunesearch_papers
Free to start. No credit card required.
Terminal
Install the CLIlune papers get 94fa7c6b-da82-4efb-90c7-e1b803162222Related papers
- LiDO-DAG: A Framework for Verifying Safety and Liveness of DAG-Based Consensus ProtocolsLongfei Qiu, Jingqi Xiao, Ji-Yong Shin, Zhong ShaoPLDI 2025 · 3 citations
- A Semi-Asynchronous Decentralized Federated Learning Framework via Tree-Graph BlockchainCheng Zhang, Yang Xu, Xiaowei Wu, En Wang et al.INFOCOM 2024 · 12 citations
- Lemonshark: Asynchronous DAG-BFT With Early FinalityMichael Yiqing Hu, Alvin Hong Yao Yan, Yihan Yang, Xiang Liu et al.NSDI 2026 · 1 citation
- Ladder: A Convergence-based Structured DAG Blockchain for High Throughput and Low LatencyDengcheng Hu, Jianrong Wang, Xiulong Liu, Hao Xu et al.NSDI 2025 · 15 citations
- Bullshark: DAG BFT Protocols Made PracticalAlexander Spiegelman, Neil Giridharan, Alberto Sonnino, Lefteris Kokoris-KogiasCCS 2022 · 132 citations
