Many words have been spilled about distributed systems of many different sizes, scales, and complexities. Most of the time though, what we talk about are generalities: algorithms and techniques. Gossip, failure detection, and leader election are discussed as if they’re a magical salve that can be rubbed onto a project to make it distributed. But the practical realities of distributed systems are far from straightforward.
Tyler McMullen outlines a new approach built to perform very high volumes of health checks across a cluster of machines for reliability and scalability. The system is complex enough that we need to worry about things like gossip, failure detection, leader election, logical clocks, and distributed key-value components. Tyler walks you through each of the major components, demonstrating how they are practically built and discussing the pain and compromises that each bring.
You’ll leave with a better understanding of how the basic components fit together in a real-world system and be better equipped to approach such projects yourself with confidence.
Tyler McMullen is CTO of Fastly, where he is responsible for the system architecture and leads the company’s technology vision. As part of the founding team, Tyler built the first versions of Fastly’s instant purging system, API, and real-time analytics. Previously, Tyler worked on text analysis and recommendations at Scribd. A self-described technology curmudgeon, Tyler has experience in everything from web design to kernel development and loathes all of it. Especially distributed systems.
©2017, 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. • email@example.com