Site Reliability Engineering

Chapter 32. The Evolving SRE Engagement Model

Written by Acacio Cruz and Ashish Bhambhani

Edited by Betsy Beyer and Tim Harvey

SRE Engagement: What, How, and Why

We’ve discussed in most of the rest of this book what happens when SRE is already in charge of a service. Few services begin their lifecycle enjoying SRE support, so there needs to be a process for evaluating a service, making sure that it merits SRE support, negotiating how to improve any deficits that bar SRE support, and actually instituting SRE support. We call this process onboarding. If you are in an environment where you are surrounded by a lot of existing services in varying states of perfection, your SRE team will probably be running through a prioritized queue of onboardings for quite a while until the team has finished taking on the highest-value targets.

Although this is very common, and a completely reasonable way of dealing with a fait accompli environment, there are actually at least two better ways of bringing the wisdom of production, and SRE support, to services old and new alike.

In the first case, just as in software engineering—where the earlier the bug is found, the cheaper it is to fix—the earlier an SRE team consultation happens, the better the service will be and the quicker it will feel the benefit. When SRE is engaged during the earliest stages of design, the time to onboard is lowered and the service is more reliable “out of the gate,” usually because we don’t have to spend the time unwinding ...