Drupal can be used to build a startup that scales, or as a critical foundation for an enterprise-grade public-facing web application.

Patterns that follow the "separation of concern" provided by microservices and decoupled Drupal become important when considering best practices for growing/maintaining applications at scale.

The "majestic monolith" architecture that Drupal fits provides much of the functionality that would take longer to build with only microservices and custom code, Drupal handles functionality such as logging, user management, and authentication all out-of-the-box in one centrally located codebase. However Drupal also offers the ability to migrate this functionality to microservices as you scale, for example porting users from Drupal to Amazon Cognito.

On the hosting and DevOps side, We will also consider SAAS/PAAS DevOps platforms and compare them against cloud services such as AWS, Azure and GCP. In what cases are these platforms the best choice for your bottom line and when is moving to a more complex DevOps pipeline based on the custom configuration of microservices reasonable?

We will also consider when it would be a good choice to forgo Drupal completely, using a decoupled frontend that relies completely on a microservices backend. We will cover the experiences we have encountered here on various projects and highlight the unexpected investment that orchestrating these various systems can incur.