.png){kind=icon}
.png){kind=icon}
Return
Chained HotStuff Under Performance Attack
.png){kind=icon}
PRE
AI
DOI：10.1109/TDSC.2025.3540980.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Chained HotStuff is a state-of-the-art Byzantine fault-tolerant protocol for building decentralized systems like blockchains. Although chained HotStuff has been widely adopted in many systems, its performance (e.g., throughput and latency) under attacks is still under-explored. In this paper, we develop a multi-metric evaluation framework to quantitatively analyze the performance of chained HotStuff with respect to its chain growth rate, chain quality, and latency. We propose several new attack strategies and evaluate their effects on the performance of chained HotStuff. Our analysis shows that the chain growth rate (resp, chain quality) of chained HotStuff under our attacks can drop to <inline-formula><tex-math notation="LaTeX">$4/9$</tex-math></inline-formula> (resp, <inline-formula><tex-math notation="LaTeX">$12/17$</tex-math></inline-formula>) of that without attacks when one-third of nodes are Byzantine. In addition, we use our framework to evaluate a variant of chained HotStuff, DiemBFT and find that some engineering optimizations render it more vulnerable to some attacks than the original chained HotStuff. Finally, we provide two countermeasures, i.e., broadcasting QCs and the longest chain rule, to thwart these attacks. Our analysis shows that the proposed countermeasures can significantly reduce the latency (almost half of that in chained HotStuff) and make it impossible for an attacker to lower the chain quality by simple attacks.
Keywords:
Byzantine fault tolerance
blockchains
chained hotstuff
forking attack
delay attack
chain quality
chain growth
Journal
IF:
7.5
Papers:
2.4K
Citations:
9.6K