In most edge networks, especially those with global presence, observability is key. While application-level metrics are a minimum bar, the addition of smart traffic analysis can often provide additional telemetry, resulting in critical insight, improving operations, and even prompting new product features. Gaining such insight, of course, comes with an array of associated challenges.
Shannon Weyrick and Clare Gollnick discuss the types of interesting data available through traffic analysis, along with strategies and technologies for a fully distributed collection and analysis framework, including fast network analysis at the edge; policy-based filtering, allowing for granular control over collection; centralized control plane; centralized collection, storage, and searching; useful aggregations such as top k and HyperLogLog; and recent advances in the Linux networking stack. All discussion is grounded in real-world use cases and examples.
Shannon Weyrick is vice president of architecture at NS1. A 20-year veteran of internet infrastructure, Shannon is an accomplished technical architect, developer, and leader whose experience encompasses both development and operations of globally distributed platforms. Previously, Shannon worked at INAP and F5. A regular open source contributor, he has led and worked on a wide range of infrastructure projects from high-performance servers to novel programming languages and runtimes, and he enjoys writing and speaking at industry conferences.
Clare Gollnick is the director of data science at NS1, an industry-leading DNS and traffic management platform. An expert on statistical inference and machine learning, Clare writes and speaks often on the intersection of data, philosophy, and entrepreneurship. Previously, as chief technology officer of Terbium Labs, Clare led a diverse team of engineers and data scientists. Her team built innovate information security products, preventing fraud while still protecting consumer privacy. Clare has published a number of academic papers on information processing within neural networks, validation of new statistical methods, and the philosophy of science. Clare holds a PhD from Georgia Tech and a BS from UC Berkeley.
©2019, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org