Designing for Scalability with Erlang/OTP

Chapter 3. Behaviors

As a prelude to learning how to structure our process supervision trees and architect our concurrency models, let’s spend some time understanding the underlying principles behind behaviors. Instead of diving straight into the world of interface functions and callbacks, we explain what goes on behind the scenes, ensuring you use OTP behaviors efficiently and understand their benefits and advantages. So, what are they?

Erlang processes that solve radically different tasks follow similar design patterns. The most commonly used patterns have been abstracted and implemented in a set of generic library modules called the OTP behaviors. When reading about behaviors, you should see them as a formalization of process design patterns.

Although the strict concept of design patterns used in object-oriented programming hasn’t been applied to Erlang, OTP provides a powerful, reusable solution for concurrent processes that hides and abstracts away all of the tricky aspects and borderline conditions. It ensures that projects do not have to reinvent the wheel, while maximizing reusability and maintainability through a solid, well-tested, generic, and reusable code base. These behaviors are, in “design pattern speak,” implementation libraries of the concurrency models.

Process Skeletons

If you try to picture an Erlang process managing a key-value store and a process responsible for managing the window of a complex GUI system, they might at first glance appear very different ...