Please log in to watch this conference skillscast.
BritishGas adopted microservices 2 years ago. A heavily distributed architectural style that comes with opportunities and challenges when it comes to understanding how a given system actually works. Mapping the dependencies between the ever growing list of involved components is becoming crucial to remain able to manage such a system. At a high level: Web journeys on the company website consume Rest endpoints, those endpoints are exposed by our microservices, microservices depend on other microservices, databases and third party APIs.
During this talk you will learn:
How to keep a self-actualising map of all these entities and how they depend on each other.
How to best expose and leverage such a map to improve operations, design and refactoring.
BritishGas approach takes data from both automated static analysis tools and live tracing & traffic sources; all combined to generate and keep an up-to-date graph of components and dependencies. A neo4j graph then used by other tools to, for example:
automatically list the endpoints and journeys impacted when a third party system is down, and notify the relevant teams,
help with bug investigations and assist in understanding the root cause of multiple problems,
list what consumers, front-end and back-end, will need re-testing when an endpoint is being refactored,
provide clear maps and graphs to architects, managers and developers, promoting better communication between teams,
detect certain duplications, coupling and other anti-pattern issues, helping toward a better architecture.
YOU MAY ALSO LIKE:
Lightning Talk: Microservices and Dependency Mapping - How to Obtain and Leverage the Graph of the System Component Interactions
Mikael is a Senior Developer at British Gas. Backend & Frontend, Microservices, APIs, AWS, iOS/Mobile apps, Architecture, Technology enthusiast. Twitter: @Mikael_wozniak