Netflix Hystrix is a well-known OSS framework for Circuit Breaker in Java. It is lesser known for its application in operational monitoring and debugging production issues, which is one of the biggest usages of Hystrix inside Netflix, where Hystrix metrics are widely used by edge services and dependent microservices to monitor dependencies’ health (including network/proxy issues and trending). When there is a production issue, this data is invaluable in isolating the problem area (within a few minutes) and expediting the recovery process.
Billy Yuen explains how Netflix applies Hystrix to existing applications with undocumented dependencies and implements DevOps monitoring with Hystrix Metrics. Billy also outlines the anti-patterns learned from implementing Hystrix.
Billy Yuen is a principal engineer in the Intuit’s Platform team, focusing on AWS and Kubernetes adoption, system resiliency, and monitoring. Previously, Billy worked on Netflix’s Edge Services team to build the next generation of edge-service infrastructure to support millions of customers (more than 3 billion requests per day) with high scalability, resilience to failure, and rapid innovation. Billy was a speaker at Java One 2016 and Velocity NY 2016 on “Operational Excellence with Netflix Hystrix”, “CI/CD at lightspeed” at KubeCon 2018, and “Automated canary release” in Container World 2019. He is also a co-author of Manning’s GitOps and Kubernetes
©2016, 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