Scholarlink

<p>In Hyperledger Fabric (HLF), Endorsement Policies (EPs) can be tailored for each application by specifying which endorsing peers from participating organisations must approve a transaction. Such customisation is essential to meet the specific security and business requirements of an application. This study implements three EPs configurations within an ongoing blockchain-based platform [Alhabib, R., Yadav, P.&#xa0;In: Proceedings of the WINCOM Conference (2024)], designed for sharing Autonomous Vehicle (AV) data. Our findings reveal a critical trade-off between achieving higher transaction success rates and maintaining optimal system throughput and latency. Specifically, more complex endorsement policies enhance fault tolerance and success rates under concurrent workloads but introduce additional computational overhead that can decrease throughput and increase latency. Conversely, simpler policies yield better throughput and lower latency but at the cost of reduced resilience and a higher likelihood of transaction failures under high user concurrency. These results underscore the importance of carefully balancing endorsement policy complexity with application-specific requirements to optimise security, fault tolerance, and performance in AV data sharing environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Evaluating the impact of endorsement policies on hyperledger fabric performance for autonomous vehicle data sharing

  • 期刊论文

  • 发表日期：

    2025年11月11日

查看全文
  • Reem Alhabib,
  • Poonam Yadav

摘要

In Hyperledger Fabric (HLF), Endorsement Policies (EPs) can be tailored for each application by specifying which endorsing peers from participating organisations must approve a transaction. Such customisation is essential to meet the specific security and business requirements of an application. This study implements three EPs configurations within an ongoing blockchain-based platform [Alhabib, R., Yadav, P. In: Proceedings of the WINCOM Conference (2024)], designed for sharing Autonomous Vehicle (AV) data. Our findings reveal a critical trade-off between achieving higher transaction success rates and maintaining optimal system throughput and latency. Specifically, more complex endorsement policies enhance fault tolerance and success rates under concurrent workloads but introduce additional computational overhead that can decrease throughput and increase latency. Conversely, simpler policies yield better throughput and lower latency but at the cost of reduced resilience and a higher likelihood of transaction failures under high user concurrency. These results underscore the importance of carefully balancing endorsement policy complexity with application-specific requirements to optimise security, fault tolerance, and performance in AV data sharing environments.