mirror of
https://github.com/lux-apps/papers.git
synced 2026-10-30 18:15:21 +00:00
zeekay
fc9b2e03e1
Auto-fixes from the paper accuracy swarm: unambiguous build-breakers, broken \ref/\cite/\input targets, and prose typos. Substantive accuracy findings (dates, contradictory numbers, LP numbering) reported separately for review.
47 lines
1.2 KiB
TeX
47 lines
1.2 KiB
TeX
\section{Benchmarks}
|
|
|
|
|
|
\subsection{Throughput}
|
|
|
|
\begin{table}[h]
|
|
\centering
|
|
\begin{tabular}{lrrr}
|
|
\toprule
|
|
\textbf{Configuration} & \textbf{TPS} & \textbf{Finality (s)} & \textbf{Bandwidth (MB/s)} \\
|
|
\midrule
|
|
ECDSA baseline & 65,000 & 1.8 & 16.7 \\
|
|
Hybrid mode & 55,000 & 1.9 & 42.3 \\
|
|
Pure ML-DSA & 50,000 & 1.95 & 33.6 \\
|
|
With aggregation & 62,000 & 1.85 & 18.9 \\
|
|
\bottomrule
|
|
\end{tabular}
|
|
\caption{Performance comparison: ECDSA vs Quantum consensus}
|
|
\end{table}
|
|
|
|
\textbf{Key findings}:
|
|
\begin{itemize}
|
|
\item 23\% TPS reduction in pure ML-DSA mode
|
|
\item Aggregation recovers 95\% of baseline performance
|
|
\item Finality time increase: Only 150ms (+8.3\%)
|
|
\end{itemize}
|
|
|
|
\subsection{Security Levels}
|
|
|
|
\begin{table}[h]
|
|
\centering
|
|
\begin{tabular}{lrr}
|
|
\toprule
|
|
\textbf{Attack} & \textbf{Classical (bits)} & \textbf{Quantum (bits)} \\
|
|
\midrule
|
|
ECDSA forgery & 128 & 64 (Grover) \\
|
|
ML-DSA forgery & 128 & 128 (MLWE) \\
|
|
Block reversal & 128 & 128 \\
|
|
Double-spend & $\infty$ (finality) & $\infty$ (finality) \\
|
|
\bottomrule
|
|
\end{tabular}
|
|
\caption{Security levels against classical and quantum adversaries}
|
|
\end{table}
|
|
|
|
\textbf{Conclusion}: ML-DSA maintains 128-bit security against quantum adversaries.
|
|
|