Jetpack: Consensus Made Generally Fast
Ze Tang, Zihao Zhang, Weihai Shen, Jicheng Shi, Shuai Mu
Abstract
Classic consensus protocols such as Raft require 2 round- trip times (RTTs) for a client to commit a command. Recent protocols reduce this latency to 1 RTT through a carefully designed fast path, but these fast paths are tightly coupled to their host protocols and cannot be retrofitted to existing systems. This paper presents Jetpack, a plugin framework that adds a 1-RTT fast path to existing consensus protocols with minimal modifications. Jetpack issues commands concurrently to both a fast path and the original path, committing when the fast path succeeds while ensuring the original path honors the same decision. The key challenge is maintaining this guarantee across view changes: we identify the view change hazard, a subtle correctness issue where promises made during stable operation can become invalid after leader elections. Jetpack captures this in two structural requirements and two design principles that a fast-path layer can apply across consensus environments, conditions that are easy to overlook when a fast-path idea is adapted to a new setting, as we illustrate with prior designs and a deployed system. We evaluate Jetpack on six consensus systems across 10 AWS datacenters. Results show that Jetpack achieves 1-RTT latency, reducing average commit latency by as much as 60%, while preserving the original path’s performance and proper- ties when the fast path is not used.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext c5ed76d1-4822-4bb0-9998-a5b094fd143fBuilds on19
- Microsecond Consensus for Microsecond ApplicationsMarcos K. Aguilera, Naama Ben-David, Rachid Guerraoui, Virendra J. Marathe et al.OSDI 2020 · 73 citations
- Kauri: Scalable BFT Consensus with Pipelined Tree-Based Dissemination and AggregationRay Neiheiser, Miguel Matos, Luís E. T. RodriguesSOSP 2021 · 65 citations
- Virtual Consensus in DelosMahesh Balakrishnan, Jason Flinn, Chen Shen, Mihir Dharamshi et al.OSDI 2020 · 42 citations
- Fault-Tolerant Replication with Pull-Based Consensus in MongoDBSiyuan Zhou, Shuai MuNSDI 2021 · 38 citations
- Gryff: Unifying Consensus and Shared RegistersMatthew Burke, Audrey Cheng, Wyatt LloydNSDI 2020 · 29 citations
Related papers
- Integrating 2PC with Consensus for Fast ReplicationYan Chen, Xinyi Yu, Shengyun Liu, Ruofan Xiong et al.SIGCOMM 2026
- CD-Raft: Reducing the Latency of Distributed Consensus in Cross-Domain SitesYangyang Wang, Ziqian Cheng, Yucong Dong, Zichen XuINFOCOM 2026
- QuePaxa: Escaping the tyranny of timeouts in consensusPasindu Tennage, Cristina Basescu, Lefteris Kokoris-Kogias, Ewa Syta et al.SOSP 2023 · 7 citations
- Chitu: Avoiding Unnecessary Fallback in Byzantine ConsensusRongji Huang, Xiangzhe Wang, Xiaofeng Yan, Lei Fan et al.USENIX ATC 2025 · 3 citations
- HovercRaft: achieving scalability and fault-tolerance for microsecond-scale datacenter servicesMarios Kogias, Edouard BugnionEuroSys 2020 · 52 citations
