Zhaoshuo Li , Liang Cai , Weiwei Qiu , Fanglei Huang , Hao Duan , Chao Yuan
{"title":"Cox: a reliable runtime switching protocol for BFT consensus algorithms","authors":"Zhaoshuo Li , Liang Cai , Weiwei Qiu , Fanglei Huang , Hao Duan , Chao Yuan","doi":"10.1016/j.bcra.2025.100345","DOIUrl":null,"url":null,"abstract":"<div><div>Different Byzantine fault tolerance (BFT) consensus algorithms have distinct characteristics and optimal use cases. Rarely does a single protocol demonstrate outstanding performance across different environments. Sometimes, systems need to switch consensus algorithms to adapt to different environments. This paper aims to investigate how to achieve seamless consensus algorithm switching without system downtime or compromising the security and consistency of the original consensus module. To address this issue, we propose Cox, a reliable online switching protocol for BFT consensus algorithms, and design a corresponding recovery sub-protocol for lagging nodes. Experimental results indicate that, under the assumption of a partially synchronous network, Cox ensures both the liveness and safety of the consensus module and achieves reliable switching in milliseconds.</div></div>","PeriodicalId":53141,"journal":{"name":"Blockchain-Research and Applications","volume":"7 3","pages":"Article 100345"},"PeriodicalIF":6.5000,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Blockchain-Research and Applications","FirstCategoryId":"1093","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2096720925000727","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/7/23 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}
引用次数: 0
Abstract
Different Byzantine fault tolerance (BFT) consensus algorithms have distinct characteristics and optimal use cases. Rarely does a single protocol demonstrate outstanding performance across different environments. Sometimes, systems need to switch consensus algorithms to adapt to different environments. This paper aims to investigate how to achieve seamless consensus algorithm switching without system downtime or compromising the security and consistency of the original consensus module. To address this issue, we propose Cox, a reliable online switching protocol for BFT consensus algorithms, and design a corresponding recovery sub-protocol for lagging nodes. Experimental results indicate that, under the assumption of a partially synchronous network, Cox ensures both the liveness and safety of the consensus module and achieves reliable switching in milliseconds.
期刊介绍:
Blockchain: Research and Applications is an international, peer reviewed journal for researchers, engineers, and practitioners to present the latest advances and innovations in blockchain research. The journal publishes theoretical and applied papers in established and emerging areas of blockchain research to shape the future of blockchain technology.