mirror of
https://github.com/hanzoai/papers.git
synced 2026-10-29 01:14:44 +00:00
zeekay
0d1d121451
Fix shared/ include paths to ../shared/ for 55 subdirectory papers so
that hanzocover.sty and lstlang.tex resolve when building from each
paper's own directory (the build pattern used by latexmk per paper).
Build fixes:
- hanzo-operative/sections/03-screen-understanding-pipeline.tex: move
\COMMENT from after \FOR{...} onto the \STATE line so the algorithmic
env doesn't mis-parse.
- hanzo-network-architecture (abstract + sections 04, 12, 13): escape
unicode mu (U+03BC) as $\mu$ for pdflatex.
- ntt-transform/sections/08-benchmark-results.tex: same mu escape.
All 55 Tier-2/3 papers now build clean with latexmk -pdf.
51 lines
1.8 KiB
TeX
51 lines
1.8 KiB
TeX
\documentclass[11pt,twocolumn]{article}
|
|
\usepackage{../shared/hanzocover}
|
|
|
|
\usepackage[utf8]{inputenc}
|
|
\usepackage[T1]{fontenc}
|
|
\usepackage{amsmath,amssymb}
|
|
\usepackage{graphicx}
|
|
\usepackage{hyperref}
|
|
\usepackage{cite}
|
|
\usepackage{algorithm}
|
|
\usepackage{algorithmic}
|
|
\usepackage{listings}
|
|
\input{../shared/lstlang}
|
|
\usepackage{xcolor}
|
|
|
|
\lstset{
|
|
basicstyle=\ttfamily\small,
|
|
keywordstyle=\color{blue},
|
|
commentstyle=\color{gray},
|
|
stringstyle=\color{red},
|
|
breaklines=true,
|
|
frame=single
|
|
}
|
|
|
|
\title{Verus: Real-Time Commerce Analytics at Scale}
|
|
|
|
\author{Zach Kelling\\
|
|
Hanzo Industries\\
|
|
\texttt{[email protected]}}
|
|
|
|
\date{November 2015}
|
|
|
|
\begin{document}
|
|
|
|
\hanzocoverpage
|
|
|
|
\begin{abstract}
|
|
We present Verus, a real-time analytics system designed for e-commerce workloads. Traditional analytics architectures batch-process events with multi-hour latency, limiting merchants' ability to respond to emerging trends, detect anomalies, and optimize campaigns in flight. Verus processes commerce events with sub-second latency while maintaining the analytical flexibility of batch systems. The architecture combines stream processing for real-time aggregation with columnar storage for ad-hoc queries, unified through a SQL interface that abstracts the underlying complexity. We describe the system design, evaluate performance on production workloads, and demonstrate applications including live dashboards, cohort analysis, and anomaly detection. Production deployments process over 50,000 events per second with query latency under 200ms for common dashboard queries.
|
|
\end{abstract}
|
|
|
|
\input{sections/01-introduction}
|
|
\input{sections/02-background}
|
|
\input{sections/03-system-design}
|
|
\input{sections/04-analytical-capabilities}
|
|
\input{sections/05-evaluation}
|
|
\input{sections/06-case-studies}
|
|
\input{sections/07-related-work}
|
|
\input{sections/99-conclusion}
|
|
|
|
\end{document}
|