mirror of
https://github.com/lux-apps/papers.git
synced 2026-10-30 18:15:21 +00:00
zeekay
b55171116e
Retract the fabricated '581M/597ns / 100M+/s / 434M GPU / planet-scale' headline (a time.Sleep sim + x40 fudge that measured nothing) across the corpus. Flagship lux-lightspeed-dex rewritten (decomplected d-chain engine / dexvm atomic proxy / Go wrapper / 0x9010 CLOB facade; measured GB10 104-126M & M4 Max ~250M fills/s byte-parity-gated; Knuth-D division lever; Block-STM settlement; P3Q->starkfri rollup to Z-Chain). Technical papers (precompile-engineering, gpu-crypto-stack, evmgpu-benchmark, p3q-corona, smart-order-routing, fhe-mpc-hybrid) carry the measured impl + cross-ref the flagship. >1B/s framed honestly as N-node aggregation, never single-GPU. All compile.
47 lines
1.1 KiB
TeX
47 lines
1.1 KiB
TeX
\section{Hardware and Environment}
|
|
|
|
|
|
All benchmarks were collected on a single machine:
|
|
|
|
\begin{table}[h]
|
|
\centering
|
|
\small
|
|
\begin{tabular}{ll}
|
|
\toprule
|
|
\textbf{Component} & \textbf{Specification} \\
|
|
\midrule
|
|
CPU & Apple M1 Max, 10 cores (8P + 2E) \\
|
|
GPU & Apple M1 Max, 32-core Metal GPU \\
|
|
RAM & 64\,GB unified memory (shared CPU/GPU) \\
|
|
OS & macOS, Darwin kernel \\
|
|
Storage & NVMe SSD \\
|
|
Go & 1.22.x \\
|
|
Clang & Apple LLVM 15.x \\
|
|
Metal & Metal 3.0, MSL 3.0 \\
|
|
\bottomrule
|
|
\end{tabular}
|
|
\caption{Benchmark hardware for the EVM, C++ interpreter, and
|
|
cryptographic benchmarks. Unified memory eliminates CPU--GPU copy
|
|
overhead.}
|
|
\label{tab:hardware}
|
|
\end{table}
|
|
|
|
The GPU CLOB matcher benchmarks (Section~\ref{sec:clob}) were collected
|
|
on two additional hosts, reported in their own table there:
|
|
|
|
\begin{table}[h]
|
|
\centering
|
|
\small
|
|
\begin{tabular}{ll}
|
|
\toprule
|
|
\textbf{Host} & \textbf{GPU} \\
|
|
\midrule
|
|
GB10 & NVIDIA Blackwell (GB10), CUDA \\
|
|
M4 Max & Apple M4 Max, Metal, unified memory \\
|
|
\bottomrule
|
|
\end{tabular}
|
|
\caption{Additional hosts used for the GPU CLOB matcher benchmarks.}
|
|
\label{tab:hardware-clob}
|
|
\end{table}
|
|
|