基于 GraphBLAS 超稀疏矩阵与 D4M 关联数组的实时网络流量分析部署
网络与互联网体系结构
2023-12-29 v2 社会与信息网络
摘要
网络的矩阵/数组分析可为其行为提供重要洞察，并有助于网络的运行与防护。先前工作已展示了 GraphBLAS（graphblas.org）超稀疏矩阵与 D4M（d4m.mit.edu）关联数组（矩阵的数学超集）在分析、性能和压缩方面的能力。要获得这些能力的益处，需将其集成到运行系统中，这带来了其特有的挑战。本文描述了两个实时运行实现的示例。其一是运行态 GraphBLAS 实现，在高带宽网络分流器上构建匿名化超稀疏矩阵。其二是运行态 D4M 实现，用于分析每日云网关日志。文中给出了这些实现的架构。收集并详细测量了资源与性能数据并进行分析。这些实现能够以 modest 的计算资源（数个处理核）满足其运行需求。GraphBLAS 非常适合于具有相对结构化网络数据的高带宽连接的底层分析。D4M 非常适合于更非结构化数据的高层分析。本工作表明这些技术可在运行环境中实现。
引用
@article{arxiv.2309.02464,
title = {Deployment of Real-Time Network Traffic Analysis using GraphBLAS Hypersparse Matrices and D4M Associative Arrays},
author = {Michael Jones and Jeremy Kepner and Andrew Prout and Timothy Davis and William Arcand and David Bestor and William Bergeron and Chansup Byun and Vijay Gadepally and Micheal Houle and Matthew Hubbell and Hayden Jananthan and Anna Klein and Lauren Milechin and Guillermo Morales and Julie Mullen and Ritesh Patel and Sandeep Pisharody and Albert Reuther and Antonio Rosa and Siddharth Samsi and Charles Yee and Peter Michaleas},
journal= {arXiv preprint arXiv:2309.02464},
year = {2023}
}
备注
Accepted to IEEE HPEC, 8 pages, 8 figures, 1 table, 69 references. arXiv admin note: text overlap with arXiv:2203.13934. text overlap with arXiv:2309.01806