ElasticDAG: Elastic DAG-based Blockchain
Author:
Affiliation:

Clc Number:

TP393

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    A directed acyclic graph (DAG)-based blockchain adopts a parallel topology and can significantly improve system performance compared with conventional chain-based blockchains with a serial topology. As a result, it has attracted wide attention from the industry. However, the storage model and the consensus protocol of the existing DAG-based blockchains are highly coupled, which lacks the flexibility to meet diversified application demands. Furthermore, most DAG-based blockchains lack flexibility at the consensus protocol level and are limited to probabilistic consensus protocols, which is difficult to take into account confirmation latency and security and is especially unfriendly to delay-sensitive applications. Therefore, this study presents the elastic DAG-based blockchain, namely ElasticDAG. The core idea is to decouple the storage model and the consensus protocol, enabling them to proceed in parallel and independently, so as to flexibly adapt to diversified applications. In order to improve the throughput and activity of the system, an adaptive block confirmation strategy and an epoch-based block ordering algorithm are designed for the storage model. In response to the need to reduce transaction confirmation latency, a low-latency DAG blockchain hybrid consensus protocol is designed. Experimental results demonstrate that the ElasticDAG prototype in WAN can achieve a throughput exceeding 11 Mb/s, and it yields a confirmation latency of tens of seconds. Compared with OHIE and Haootia, ElasticDAG can reduce confirmation latency by 17 times and improve security from 91.04% to 99.999 914% while maintaining the same throughput and consensus latency.

    Reference
    Related
    Cited by
Get Citation

岳镜涛,肖江,张世桀,程凤,陈汉华,金海. ElasticDAG: 弹性图式区块链.软件学报,2024,35(11):5279-5305

Copy
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:July 30,2022
  • Revised:April 28,2023
  • Adopted:
  • Online: January 24,2024
  • Published:
You are the firstVisitors
Copyright: Institute of Software, Chinese Academy of Sciences Beijing ICP No. 05046678-4
Address:4# South Fourth Street, Zhong Guan Cun, Beijing 100190,Postal Code:100190
Phone:010-62562563 Fax:010-62562533 Email:jos@iscas.ac.cn
Technical Support:Beijing Qinyun Technology Development Co., Ltd.

Beijing Public Network Security No. 11040202500063