Microservices: So Crazy It Just Might Work
Event Info
Description
Want to eliminate technical debt, deploy every few minutes, use Python, Clojure, and .NET in the same system, and make major rewrites trivially easy? Advocates of microservices designs, such as Martin Fowler, claim that you can do all this and much more if you construct your application as a collection of many tiny components connected via a messaging mechanism. Join Douglas Squirrel for a critical analysis of these claims!
We'll review the brief history of microservices and discuss of the pros and cons of this design style. We will ask questions like: * What are the historical antecedents of microservices architectures and why are they coming to the fore now? * What is the value of unit testing when your units are independent services implemented in 100 lines of code or less? * How might you manage risks in a microservices system, such as unwanted emergent behaviour and component failure? * What might a microservices development team look like? What skills should team members have (or develop) and what roles will they play?
Bring a laptop running a compiler or interpreter for your favourite language if you'd like to try building a working component in a microservices system.
