Files
zeekay abf99d05fc papers: reorganize into per-paper dirs, add Antje Worring as co-author
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.
2026-04-18 21:50:07 -07:00

64 lines
2.0 KiB
TeX

\section{Experimental Setup}
%% ============================================================================
\subsection{Hardware}
All experiments use identical bare-metal nodes:
\begin{table}[h]
\centering
\begin{tabular}{@{}ll@{}}
\toprule
\textbf{Component} & \textbf{Specification} \\
\midrule
CPU & AMD EPYC 9454 (48 cores, 2.75\,GHz) \\
RAM & 256\,GB DDR5-4800 ECC \\
Storage & 2$\times$ 3.84\,TB NVMe Gen5 (RAID~0) \\
Network & 25\,Gbps bonded Ethernet \\
OS & Ubuntu 24.04 LTS, kernel 6.8 \\
\bottomrule
\end{tabular}
\caption{Node hardware specification.}
\label{tab:hardware}
\end{table}
\subsection{Geographic Distribution}
Three datacenters provide realistic WAN conditions:
\begin{table}[h]
\centering
\begin{tabular}{@{}llll@{}}
\toprule
\textbf{Site} & \textbf{Location} & \textbf{RTT to SFO} & \textbf{RTT to NYC} \\
\midrule
SFO & San Francisco, CA & --- & 62\,ms \\
NYC & New York, NY & 62\,ms & --- \\
MCI & Kansas City, MO & 38\,ms & 28\,ms \\
\bottomrule
\end{tabular}
\caption{Inter-datacenter round-trip latencies (measured via ICMP, median of 10{,}000 samples).}
\label{tab:latency}
\end{table}
Validators are distributed equally across all three sites (e.g., 7/7/7 for $n=21$,
34/33/33 for $n=100$).
\subsection{Methodology}
\begin{itemize}
\item \textbf{Warm-up}: 60\,s of sustained load before measurement begins.
\item \textbf{Duration}: Each test runs for 300\,s (5\,min) of steady-state operation.
\item \textbf{Repetitions}: 10 independent runs per configuration. We report median
and p99 values.
\item \textbf{Load generation}: Dedicated client nodes (not validators) inject
transactions at increasing rates until saturation.
\item \textbf{Finality measurement}: Wall-clock time from transaction submission to
irreversible commitment, measured at the client.
\item \textbf{Software versions}: Quasar v1.4.0, CometBFT v0.38.15, aptos-core v1.18,
sui v1.36, Lighthouse v5.3, Agave v2.0, AvalancheGo v1.11.12.
\end{itemize}
%% ============================================================================