Characterizing the Impact of Network Delay on Bitcoin Mining

Tong Cao, Jérémie Decouchant, Jiangshan Yu, Paulo Esteves-Verissimo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Scopus citations

Abstract

While previous works have discussed the network delay upper bound that guarantees the consistency of Nakamoto consensus, measuring the actual network latencies and evaluating their impact on miners/pools in Bitcoin remain open questions. This paper fills this gap by: (1) defining metrics that quantify the impact of network latency on the mining network; (2) developing a tool, named miner entanglement (ME), to experimentally evaluate these metrics with a focus on the network latency of the top mining pools; and (3) quantifying the impact of the current network delays on Bitcoin's mining network. For example, we evaluated that Poolin, a Bitcoin mining pool, was able to gain between 0.5% and 1.9% of blocks in addition (i.e., from 36.27 BTC to 137.83 BTC) per week thanks to its low network latency. Moreover, as pools are rational in Bitcoin, we model the strategy a pool would follow to improve its network latency (e.g., by leveraging our ME tool) as a two party game. We show that a Bitcoin mining pool could improve its effective hash rate by up to 4.5%. For a multi-party game, we use a state-of-the-art Bitcoin mining simulator to study the situation where all pools attempt to improve their network latency and show that the largest mining pools would improve their revenue and reach a Nash equilibrium while the smaller mining pools would suffer from a decreased access to the network, and therefore a decreased revenue. These conclusions further incentivize the centralisation of the mining network in Bitcoin, and provide an empirical explanation for the observed tendency of pools to design and rely on low latency private networks.
Original languageEnglish (US)
Title of host publication2021 40th International Symposium on Reliable Distributed Systems (SRDS)
PublisherIEEE
ISBN (Print)978-1-6654-3820-9
DOIs
StatePublished - Nov 22 2021

Bibliographical note

KAUST Repository Item: Exported on 2021-11-24
Acknowledgements: This work is partially supported by the Fonds National de la Recherche Luxembourg (FNR) through PEARL grant FNR/P14/8149128

Fingerprint

Dive into the research topics of 'Characterizing the Impact of Network Delay on Bitcoin Mining'. Together they form a unique fingerprint.

Cite this