In this workshop we’ll dive into the topic of Autonomous Service development, with a sandbox project containing several interdependent services. These projects send messages to each other, and they call each other to retrieve data. In other words, they’re not autonomous at all. We’ll figure out some ways in which we can invert dependencies and leverage a messaging solution to achieve autonomy after all.
In a relatively short period you’ll get acquainted with the basics of asynchronous integration, CQRS, domain events, and integrating bounded contexts.
There will be a healthy mixture between theory and practice. We look at different ways in which services can be dependent on each other: by directly sending queries or commands, or by knowing too much about each other's domain. We explore alternatives to improve the situation, increasing autonomy with every exercise.
Because we'll be working with an actual application, it'll be clear that there is a trade-off involved in making a service truly autonomous. We discuss the costs, the benefits, and the reasons for splitting monolithic applications into services.
This is a training for developers who have experience working on a monolithic application and are now looking for ways to split the application into multiple applications that can work together, without depending too heavily on each other.
The sandbox project we use for this training is built with PHP (without a particular framework) and requires Docker to run on your laptop. If you have experience with another object-oriented language, you should be able to do the exercises as well.