Explore an innovative approach to network fault detection in this 29-minute LISA19 conference talk. Delve into Facebook's Network Fault Finding System, which utilizes packet loss triangulation to identify misbehaving network elements. Learn how active probing with test traffic can overcome limitations of traditional monitoring methods, especially in large-scale networks. Discover the process of building a similar system using open-source tools, and witness a live demonstration on a lab network. Gain insights into the system's workflow, from introduction and context to implementation and real-time fault detection. Perfect for network engineers and administrators seeking advanced techniques for maintaining network health and performance.
Overview
Syllabus
Introduction
Context
Active probing
Requirements
Building Blocks
Master
Workflow
Demo
Outro
Taught by
USENIX