No matter your experience, at a certain point Drupal can become a bit of a black box, like any other framework. This is especially the case on a development team - original members are more familiar, and making meaningful updates that don't impact production is more difficult for junior members. How can we make certain performance stays a priority, and that each team can understand as much as possible about how the application works?
What we need is a tool that allows use to follow function calls deep within our application - a profiler. With this, we can clearly see which transactions affect performance the most, and get a clear picture of exactly what's happening when a user performs some action on our site. With that information in hand, we can make targeted and incremental steps to move it to a more performant state. The trick after, is keeping it there alongside the development lifecycle, and a hundred other features that will be added by our team over time.
In this talk, we'll investigate how profiling with Blackfire.io can provide insights into performance bottlenecks in our code, and allow teams to enforce performance expectations in the long term to protect against regressions introduced (but otherwise undetected) alongside new features.