当前位置: X-MOL 学术Future Gener. Comput. Syst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
DR-BFT: A consensus algorithm for blockchain-based multi-layer data integrity framework in dynamic edge computing system
Future Generation Computer Systems ( IF 6.2 ) Pub Date : 2021-05-11 , DOI: 10.1016/j.future.2021.04.020
Yuqi Fan , Huanyu Wu , Hye-Young Paik

Edge computing, as a part a distributed computing architecture, has become an increasingly popular paradigm. It expands the capacity of cloud by facilitating data from the end devices to be stored and processed at the edge of the network closer to the data instead of delivering it to the cloud. Data integrity is a big concern in edge computing. As a promising solution to data integrity, blockchain is viable to protect the data in edge computing systems. However, most existing consensus algorithms cannot meet the requirements of edge computing in a dynamic network, where the nodes may join or leave the blockchain network dynamically. In this paper, we introduce a two-layer blockchain-based framework to provide data integrity in edge computing, and propose a novel Dynamic Random Byzantine Fault Tolerance (DR-BFT) consensus algorithm. DR-BFT consists of three sub-algorithms, String Consensus, Data Correctness Validation, and Binary Consensus. String consensus tries to reach consensus on the data of end devices or edge servers, and the sub-algorithm is based on an agreement and borrows some ideas from the Phase King Protocol. If the string consensus fails “early termination”, each node will agree on a value from a random primary node and go through data correctness validation sub-algorithm. The system then reaches consensus on the data with binary consensus sub-algorithm which is a variant of Ben-Or and Michael’s Random Consensus. We also propose an improved quorum method to cope with contention and dynamic node leaving/joining. We analyze DR-BFT with regard to consensus correctness, security, and system overhead. DR-BFT satisfies agreement, validity and termination. We conduct experiments through simulations. Experimental results show that the proposed consensus DR-BFT can effectively improve the performance in dynamic edge computing, including communication overhead and consensus latency.



中文翻译:

DR-BFT:动态边缘计算系统中基于区块链的多层数据完整性框架的共识算法

边缘计算作为分布式计算体系结构的一部分,已成为一种越来越流行的范例。它通过促进来自终端设备的数据在网络边缘(更靠近数据)存储和处理,而不是将其传递到云来扩展云的容量。数据完整性是边缘计算中的一个大问题。作为有前途的数据完整性解决方案,区块链可以保护边缘计算系统中的数据。但是,大多数现有的共识算法无法满足动态网络中边缘计算的要求,在动态网络中节点可以动态加入或离开区块链网络。在本文中,我们介绍了一个基于两层区块链的框架来提供边缘计算中的数据完整性,并提出了一种新颖的动态随机拜占庭容错(DR-BFT)共识算法。DR-BFT由三个子算法组成:字符串共识,数据正确性验证和二进制共识。字符串共识试图在终端设备或边缘服务器的数据上达成共识,并且子算法基于协议,并且借鉴了Phase King协议的一些想法。如果字符串共识未能“提前终止”,则每个节点都将同意来自随机主节点的值,并通过数据正确性验证子算法。然后,系统使用二元共识子算法(Ben-Or和Michael's Random Consensus)的变体对数据达成共识。我们还提出了一种改进的仲裁方法来应对争用和动态节点离开/加入。我们分析DR-BFT的共识正确性,安全性和系统开销。DR-BFT满足协议,有效性和终止。我们通过模拟进行实验。实验结果表明,提出的共识DR-BFT可以有效地提高动态边缘计算的性能,包括通信开销和共识等待时间。

更新日期:2021-05-26
down
wechat
bug