International Association for Cryptologic Research

International Association
for Cryptologic Research

CryptoDB

Practical Settlement Bounds for Longest-Chain Consensus

Authors:
Peter Gaži , IOG
Ling Ren , University of Illinois at Urbana-Champaign
Alexander Russell , University of Connecticut and IOG
Download:
DOI: 10.1007/978-3-031-38557-5_4 (login may be required)
Search ePrint
Search Google
Presentation: Slides
Conference: CRYPTO 2023
Abstract: Nakamoto's longest-chain consensus paradigm now powers the bulk of the world's cryptocurrencies and distributed finance infrastructure. An emblematic property of longest-chain consensus is that it provides probabilistic settlement guarantees that strengthen over time. This makes the exact relationship between settlement error and settlement latency a critical aspect of the protocol that both users and system designers must understand to make informed decisions. A recent line of work has finally provided a satisfactory rigorous accounting of this relationship for proof-of-work longest-chain protocols, but those techniques do not appear to carry over to the proof-of-stake setting. This article develops a new analytic approach for establishing such settlement guarantees that yields explicit, rigorous settlement bounds for proof-of-stake longest-chain protocols, placing them on equal footing with their proof-of-work counterparts. Our techniques apply with some adaptations to the proof-of-work setting where they provide improvements to the state-of-the-art settlement bounds for proof-of-work protocols.
BibTeX
@inproceedings{crypto-2023-33163,
  title={Practical Settlement Bounds for Longest-Chain Consensus},
  publisher={Springer-Verlag},
  doi={10.1007/978-3-031-38557-5_4},
  author={Peter Gaži and Ling Ren and Alexander Russell},
  year=2023
}