mirror of
https://github.com/lux-apps/papers.git
synced 2026-10-30 18:15:21 +00:00
zooqueen
ebe2ebcb3c
Ringtail (Boschini et al., ePrint 2024/1113) is third-party prior art; Lux's Corona (Ring-LWE = Module-LWE rank-1) builds on it; Pulsar is Module-LWE rank-k. Retired the conflated boschini2024corona/corona2024/coronago/... bib keys to the single canonical ringtail/corona/pulsar/magnetar keys, rewrote \cite + \bibitem by context, and fixed 'Pulsar (Ring-LWE)' mislabels (Corona where corona.metal/ single-ring/Ringtail-derived; Pulsar where Module-LWE rank-k/ML-DSA). Cryptographer-led.
47 lines
2.1 KiB
TeX
47 lines
2.1 KiB
TeX
\section{Introduction}
|
|
|
|
%% ============================================================================
|
|
|
|
Consensus protocol selection determines the security ceiling, performance floor, and
|
|
operational characteristics of a blockchain network. Published benchmarks from protocol
|
|
vendors are marketing artifacts, not science. This paper provides controlled,
|
|
apples-to-apples measurements of seven consensus protocols on identical infrastructure.
|
|
|
|
Quasar~\cite{quasar-paper} is Lux Network's consensus protocol, descended from the
|
|
HotStuff lineage of leader-based BFT protocols~\cite{hotstuff}. It is \emph{not} a
|
|
Snow-family protocol. Quasar introduces dual-certificate finality---requiring both
|
|
a classical BLS aggregate signature and a post-quantum Pulsar threshold
|
|
signature~\cite{pulsar} for block finalization---while maintaining sub-second
|
|
finality and linear message complexity.
|
|
|
|
\subsection{Protocols Under Test}
|
|
|
|
\begin{table}[h]
|
|
\centering
|
|
\small
|
|
\begin{tabular}{@{}llll@{}}
|
|
\toprule
|
|
\textbf{Protocol} & \textbf{Network} & \textbf{Family} & \textbf{BFT Threshold} \\
|
|
\midrule
|
|
Quasar & Lux & Leader-based (HotStuff descendant) & $f < n/3$ \\
|
|
Tendermint & Cosmos & Leader-based (PBFT variant) & $f < n/3$ \\
|
|
HotStuff & Aptos/Diem & Leader-based (chained BFT) & $f < n/3$ \\
|
|
Narwhal+Tusk & Sui & DAG-based mempool + BFT & $f < n/3$ \\
|
|
Casper FFG & Ethereum & Checkpoint finality gadget & $f < n/3$ \\
|
|
Tower BFT & Solana & PoH-assisted BFT & $f < n/3$ \\
|
|
Snow & Avalanche & Probabilistic sampling & Probabilistic \\
|
|
\bottomrule
|
|
\end{tabular}
|
|
\caption{Consensus protocols evaluated in this study.}
|
|
\label{tab:protocols}
|
|
\end{table}
|
|
|
|
\subsection{Scope and Non-Goals}
|
|
|
|
This paper measures consensus-layer performance only. It does not evaluate execution
|
|
engines (EVM, MoveVM, SVM), mempool designs, or application-layer throughput. All
|
|
transaction payloads are 250-byte no-op transfers to isolate consensus overhead from
|
|
execution cost.
|
|
|
|
%% ============================================================================
|