$\mathsf{Aurora}$Aurora：具有高吞吐量的无领导状态机复制

IF 3.6 2区计算机科学 Q2 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

IEEE Transactions on Computers Pub Date : 2025-02-11 DOI:10.1109/TC.2025.3540656

Hao Lu;Jian Liu;Kui Ren

{"title":"$\\mathsf{Aurora}$Aurora：具有高吞吐量的无领导状态机复制","authors":"Hao Lu;Jian Liu;Kui Ren","doi":"10.1109/TC.2025.3540656","DOIUrl":null,"url":null,"abstract":"State-machine replication (SMR) allows a deterministic state machine to be replicated across a set of replicas and handle clients’ requests as a single machine. Most existing SMR protocols are leader-based requiring a leader to order requests and coordinate the protocol. This design places a disproportionately high load on the leader, inevitably impairing the scalability. If the leader fails, a complex and bug-prone fail-over protocol is needed to switch to a new leader. An adversary can also exploit the fail-over protocol to slow down the protocol. In this paper, we propose a crash-fault tolerant SMR named <inline-formula><tex-math>$\\mathsf{Aurora}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi>Aurora</mml:mi></mml:mrow></mml:math><inline-graphic></alternatives></inline-formula>, with the following properties: <list> <list-item> <label>•</label> Leaderless: it does not require a leader, hence completely get rid of the fail-over protocol.</list-item> <list-item> <label>•</label> Scalable: it can scale up to <inline-formula><tex-math>$11$</tex-math><alternatives><mml:math><mml:mn>11</mml:mn></mml:math><inline-graphic></alternatives></inline-formula> replicas.</list-item> <list-item> <label>•</label> Robust: it behaves well even under a poor network connection.</list-item></list> We provide a full-fledged implementation of <inline-formula><tex-math>$\\mathsf{Aurora}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi>Aurora</mml:mi></mml:mrow></mml:math><inline-graphic></alternatives></inline-formula> and systematically evaluate its performance. Our benchmark results show that <inline-formula><tex-math>$\\mathsf{Aurora}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi>Aurora</mml:mi></mml:mrow></mml:math><inline-graphic></alternatives></inline-formula> achieves a throughput of around two million Transactions Per Second (TPS), up to 8.7<inline-formula><tex-math>$\\boldsymbol{\\times}$</tex-math><alternatives><mml:math><mml:mo>×</mml:mo></mml:math><inline-graphic></alternatives></inline-formula> higher than the state-of-the-art leaderless SMR.","PeriodicalId":13087,"journal":{"name":"IEEE Transactions on Computers","volume":"74 5","pages":"1690-1701"},"PeriodicalIF":3.6000,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"$\\\\mathsf{Aurora}$Aurora: Leaderless State-Machine Replication With High Throughput\",\"authors\":\"Hao Lu;Jian Liu;Kui Ren\",\"doi\":\"10.1109/TC.2025.3540656\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"State-machine replication (SMR) allows a deterministic state machine to be replicated across a set of replicas and handle clients’ requests as a single machine. Most existing SMR protocols are leader-based requiring a leader to order requests and coordinate the protocol. This design places a disproportionately high load on the leader, inevitably impairing the scalability. If the leader fails, a complex and bug-prone fail-over protocol is needed to switch to a new leader. An adversary can also exploit the fail-over protocol to slow down the protocol. In this paper, we propose a crash-fault tolerant SMR named <inline-formula><tex-math>$\\\\mathsf{Aurora}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi>Aurora</mml:mi></mml:mrow></mml:math><inline-graphic></alternatives></inline-formula>, with the following properties: <list> <list-item> <label>•</label> Leaderless: it does not require a leader, hence completely get rid of the fail-over protocol.</list-item> <list-item> <label>•</label> Scalable: it can scale up to <inline-formula><tex-math>$11$</tex-math><alternatives><mml:math><mml:mn>11</mml:mn></mml:math><inline-graphic></alternatives></inline-formula> replicas.</list-item> <list-item> <label>•</label> Robust: it behaves well even under a poor network connection.</list-item></list> We provide a full-fledged implementation of <inline-formula><tex-math>$\\\\mathsf{Aurora}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi>Aurora</mml:mi></mml:mrow></mml:math><inline-graphic></alternatives></inline-formula> and systematically evaluate its performance. Our benchmark results show that <inline-formula><tex-math>$\\\\mathsf{Aurora}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi>Aurora</mml:mi></mml:mrow></mml:math><inline-graphic></alternatives></inline-formula> achieves a throughput of around two million Transactions Per Second (TPS), up to 8.7<inline-formula><tex-math>$\\\\boldsymbol{\\\\times}$</tex-math><alternatives><mml:math><mml:mo>×</mml:mo></mml:math><inline-graphic></alternatives></inline-formula> higher than the state-of-the-art leaderless SMR.\",\"PeriodicalId\":13087,\"journal\":{\"name\":\"IEEE Transactions on Computers\",\"volume\":\"74 5\",\"pages\":\"1690-1701\"},\"PeriodicalIF\":3.6000,\"publicationDate\":\"2025-02-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE Transactions on Computers\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/10880100/\",\"RegionNum\":2,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Computers","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10880100/","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}

引用次数: 0

摘要

状态机复制（SMR）允许跨一组副本复制确定性状态机，并将客户机请求作为单个机器处理。大多数现有的SMR协议都是基于leader的，需要一个leader来命令请求和协调协议。这种设计给领导者带来了不成比例的高负载，不可避免地损害了可伸缩性。如果leader发生故障，则需要一个复杂且容易出错的故障转移协议来切换到新的leader。攻击者还可以利用故障转移协议来减慢协议的速度。在本文中，我们提出了一个名为$\mathsf{Aurora}$Aurora的容错SMR，它具有以下属性：•无领导：它不需要领导者，因此完全摆脱了故障转移协议。•可扩展：它可以扩展到11美元11美元的副本。鲁棒性：即使在较差的网络连接下也表现良好。我们提供了一个完整的$\mathsf{Aurora}$Aurora的实现，并系统地评估其性能。我们的基准测试结果表明，$\mathsf{Aurora}$Aurora实现了大约每秒200万事务（TPS）的吞吐量，比最先进的无领导SMR高出8.7$\boldsymbol{\times}$ x。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

$\mathsf{Aurora}$Aurora: Leaderless State-Machine Replication With High Throughput

State-machine replication (SMR) allows a deterministic state machine to be replicated across a set of replicas and handle clients’ requests as a single machine. Most existing SMR protocols are leader-based requiring a leader to order requests and coordinate the protocol. This design places a disproportionately high load on the leader, inevitably impairing the scalability. If the leader fails, a complex and bug-prone fail-over protocol is needed to switch to a new leader. An adversary can also exploit the fail-over protocol to slow down the protocol.

In this paper, we propose a crash-fault tolerant SMR named $\mathsf{Aurora}$Aurora, with the following properties: •

Leaderless: it does not require a leader, hence completely get rid of the fail-over protocol.

•

Scalable: it can scale up to $11$11 replicas.

•

Robust: it behaves well even under a poor network connection.

We provide a full-fledged implementation of $\mathsf{Aurora}$Aurora and systematically evaluate its performance. Our benchmark results show that $\mathsf{Aurora}$Aurora achieves a throughput of around two million Transactions Per Second (TPS), up to 8.7$\boldsymbol{\times}$× higher than the state-of-the-art leaderless SMR.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

IEEE Transactions on Computers 工程技术-工程：电子与电气

CiteScore

6.60

自引率

5.40%

发文量

199

审稿时长

6.0 months

期刊介绍： The IEEE Transactions on Computers is a monthly publication with a wide distribution to researchers, developers, technical managers, and educators in the computer field. It publishes papers on research in areas of current interest to the readers. These areas include, but are not limited to, the following: a) computer organizations and architectures; b) operating systems, software systems, and communication protocols; c) real-time systems and embedded systems; d) digital devices, computer components, and interconnection networks; e) specification, design, prototyping, and testing methods and tools; f) performance, fault tolerance, reliability, security, and testability; g) case studies and experimental and theoretical evaluations; and h) new and important applications and trends.