mirror of
https://github.com/lux-apps/papers.git
synced 2026-10-30 18:15:21 +00:00
zeekay
abf99d05fc
Restructured from flat .tex/.pdf files into per-paper subdirectories. Added Antje Worring as co-author on all papers. Synced PDFs to pdfs/ for download site.
34 lines
1.1 KiB
TeX
34 lines
1.1 KiB
TeX
\section{Empirical Evaluation}
|
|
|
|
\label{sec:evaluation}
|
|
|
|
\subsection{Setup}
|
|
|
|
1,000-node testnet, pipeline depth $m = 128$, parameters $k = 20$, $\alpha = 0.8$,
|
|
$\beta = 11$, round time $\Delta = 50$ms.
|
|
|
|
\subsection{Results}
|
|
|
|
\begin{table}[h]
|
|
\centering
|
|
\begin{tabular}{@{}lrrr@{}}
|
|
\toprule
|
|
\textbf{Protocol} & \textbf{TPS} & \textbf{Latency (median)} & \textbf{Messages/tx} \\
|
|
\midrule
|
|
Sequential Wave & 90/conflict & 550ms & 220 \\
|
|
Nova (burst) & 8,500 & 400ms & 2.2 \\
|
|
Photon (pipelined) & 10,000 & 500ms & 1.8 \\
|
|
\bottomrule
|
|
\end{tabular}
|
|
\caption{Throughput comparison across consensus protocols}
|
|
\label{tab:photon-results}
|
|
\end{table}
|
|
|
|
Photon achieved 10,000 TPS---the transaction arrival rate---with 500ms median per-transaction
|
|
latency (same as non-pipelined). The key advantage is that throughput is decoupled from
|
|
per-transaction latency: increasing $\beta$ for stronger safety does not reduce throughput.
|
|
|
|
%% -----------------------------------------------------------------------
|
|
%% 8. CONCLUSION
|
|
%% -----------------------------------------------------------------------
|