An optimisation for a two‐round good‐case latency protocol
Abstract Byzantine broadcast is a fundamental primitive in distributed computing. A highly efficient Byzantine broadcast protocol, motivated by the real‐world performance of practical state machine replication protocols, is increasingly needed. This article focuses on the state‐of‐the‐art partially...
Saved in:
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2023-07-01
|
Series: | IET Information Security |
Subjects: | |
Online Access: | https://doi.org/10.1049/ise2.12123 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1832547365779668992 |
---|---|
author | Kexin Hu Zhenfeng Zhang Kaiwen Guo Weiyu Jiang Xiaoman Li Jiang Han |
author_facet | Kexin Hu Zhenfeng Zhang Kaiwen Guo Weiyu Jiang Xiaoman Li Jiang Han |
author_sort | Kexin Hu |
collection | DOAJ |
description | Abstract Byzantine broadcast is a fundamental primitive in distributed computing. A highly efficient Byzantine broadcast protocol, motivated by the real‐world performance of practical state machine replication protocols, is increasingly needed. This article focuses on the state‐of‐the‐art partially synchronous Byzantine broadcast protocol proposed by Abraham et al. (PODC’21), which achieves optimal good‐case latency of two rounds and optimal resilience of n ≥ 5f − 1 in this setting. Each step of the protocol is analysed, and then improved by cutting down the number of messages required to be collected and transmitted in the heaviest step of the protocol by about half, without adding any extra cost. This benefits from a new property, named “spread”, that we identify and extract from the original protocol. It helps us to eliminate non‐essential work in its view‐change procedure. The authors also show that no further reduction is possible without violating security. A prototype is implemented and the performances of improved and original protocols are evaluated in the same environment. The results show that our improvement can achieve about 50% lower communication cost and 40% shorter latency at a scale of 100 replicas. The latency gap becomes wider as the scale further increases. |
format | Article |
id | doaj-art-72ef722a58cb4a72a43d16347d6f7ba9 |
institution | Kabale University |
issn | 1751-8709 1751-8717 |
language | English |
publishDate | 2023-07-01 |
publisher | Wiley |
record_format | Article |
series | IET Information Security |
spelling | doaj-art-72ef722a58cb4a72a43d16347d6f7ba92025-02-03T06:45:07ZengWileyIET Information Security1751-87091751-87172023-07-0117466468010.1049/ise2.12123An optimisation for a two‐round good‐case latency protocolKexin Hu0Zhenfeng Zhang1Kaiwen Guo2Weiyu Jiang3Xiaoman Li4Jiang Han5Institute of Software Chinese Academy of Sciences Beijing ChinaInstitute of Software Chinese Academy of Sciences Beijing ChinaInstitute of Software Chinese Academy of Sciences Beijing ChinaHuawei Technologies Beijing ChinaDepartment of E‐commerce School of Management Henan University of Technology Zhengzhou ChinaInstitute of Software Chinese Academy of Sciences Beijing ChinaAbstract Byzantine broadcast is a fundamental primitive in distributed computing. A highly efficient Byzantine broadcast protocol, motivated by the real‐world performance of practical state machine replication protocols, is increasingly needed. This article focuses on the state‐of‐the‐art partially synchronous Byzantine broadcast protocol proposed by Abraham et al. (PODC’21), which achieves optimal good‐case latency of two rounds and optimal resilience of n ≥ 5f − 1 in this setting. Each step of the protocol is analysed, and then improved by cutting down the number of messages required to be collected and transmitted in the heaviest step of the protocol by about half, without adding any extra cost. This benefits from a new property, named “spread”, that we identify and extract from the original protocol. It helps us to eliminate non‐essential work in its view‐change procedure. The authors also show that no further reduction is possible without violating security. A prototype is implemented and the performances of improved and original protocols are evaluated in the same environment. The results show that our improvement can achieve about 50% lower communication cost and 40% shorter latency at a scale of 100 replicas. The latency gap becomes wider as the scale further increases.https://doi.org/10.1049/ise2.12123distributed algorithmsfault tolerance |
spellingShingle | Kexin Hu Zhenfeng Zhang Kaiwen Guo Weiyu Jiang Xiaoman Li Jiang Han An optimisation for a two‐round good‐case latency protocol IET Information Security distributed algorithms fault tolerance |
title | An optimisation for a two‐round good‐case latency protocol |
title_full | An optimisation for a two‐round good‐case latency protocol |
title_fullStr | An optimisation for a two‐round good‐case latency protocol |
title_full_unstemmed | An optimisation for a two‐round good‐case latency protocol |
title_short | An optimisation for a two‐round good‐case latency protocol |
title_sort | optimisation for a two round good case latency protocol |
topic | distributed algorithms fault tolerance |
url | https://doi.org/10.1049/ise2.12123 |
work_keys_str_mv | AT kexinhu anoptimisationforatworoundgoodcaselatencyprotocol AT zhenfengzhang anoptimisationforatworoundgoodcaselatencyprotocol AT kaiwenguo anoptimisationforatworoundgoodcaselatencyprotocol AT weiyujiang anoptimisationforatworoundgoodcaselatencyprotocol AT xiaomanli anoptimisationforatworoundgoodcaselatencyprotocol AT jianghan anoptimisationforatworoundgoodcaselatencyprotocol AT kexinhu optimisationforatworoundgoodcaselatencyprotocol AT zhenfengzhang optimisationforatworoundgoodcaselatencyprotocol AT kaiwenguo optimisationforatworoundgoodcaselatencyprotocol AT weiyujiang optimisationforatworoundgoodcaselatencyprotocol AT xiaomanli optimisationforatworoundgoodcaselatencyprotocol AT jianghan optimisationforatworoundgoodcaselatencyprotocol |