基于联盟链微电网交易的改进Raft共识算法  

Improvement of Raft consensus algorithm based on alliance chain microgrid trading

作　　者：张铭泉[1] 曹新宇 Zhang Mingquan;Cao Xinyu(School of Control&Computer Engineering,North China Electric Power University,Baoding Hebei 071003,China)

机构地区：[1]华北电力大学控制与计算机工程学院,河北保定071003

出　　处：《计算机应用研究》2024年第10期2911-2917,共7页Application Research of Computers

基　　金：中央高校基本科研业务费专项资金资助项目(2020MS122)。

摘　　要：针对联盟链微电网交易场景的高吞吐量与抵御拜占庭节点攻击的需求,提出了一种基于Raft的多领导者拜占庭容错共识算法MLB-Raft(multi-leader Byzantine fault tolerance-Raft)。首先使用可验证随机函数VRF选举领导者节点群,通过多领导者并行提交区块的方式提高算法的吞吐量;接着引入了协调者角色,负责领导者的选举、管理与系统共识;在领导者与跟随者进行区块复制的过程中,结合并简化了PBFT算法的共识流程,实现本算法的抗拜占庭特性。实验结果表明,在大规模网络节点环境下,相较于Raft算法,该算法提高了吞吐量与共识效率,但付出了部分通信开销代价;相较于PBFT算法,该算法提高了拜占庭容错能力,降低了通信开销。综上,该算法能有效保障联盟链微电网交易的时效性与安全性。To address the high throughput and resistance to Byzantine node attacks in the microgrid trading scenario of consortium chains,this paper proposed a multi-leader Byzantine fault tolerance consensus algorithm based on Raft,named MLB-Raft.The algorithm first utilized the VRF to select a group of leader nodes,increasing throughput by allowing multiple leaders to submit blocks in parallel.A coordinator role was then introduced,responsible for the election and management of leaders and system consensus.During the block replication process between leaders and followers,the consensus process of the practical Byzantine fault tolerance(PBFT)algorithm was integrated and simplified to achieve the Byzantine resilience of this algorithm.Experimental results show that,under a large-scale network node environment,compared to the Raft algorithm,this algorithm improves throughput and consensus efficiency at the cost of some communication overhead.Compared to the PBFT algorithm,this algorithm enhances Byzantine fault tolerance and reduces communication overhead.In summary,this algorithm can effectively ensure the timeliness and security of alliance chain microgrid transactions.

关 键 词：联盟链 微电网 共识算法 Raft算法 PBFT算法 可验证随机函数 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]