Although service-level objectives (SLOs) continue to grow in importance, there’s a distinct lack of information about how to implement them. Practical advice that does exist usually assumes that your team already has the infrastructure, tooling, and culture in place. In this book, recognized SLO expert Alex Hidalgo explains how to build an SLO culture from the ground up.

Ideal as a primer and daily reference for anyone creating both the culture and tooling necessary for SLO-based approaches to reliability, this guide provides detailed analysis of advanced SLO and service-level indicator (SLI) techniques. Armed with mathematical models and statistical knowledge to help you get the most out of an SLO-based approach, you’ll learn how to build systems capable of measuring meaningful SLIs with buy-in across all departments of your organization.

• Define SLIs that meaningfully measure the reliability of a service from a user’s perspective
• Choose appropriate SLO targets, including how to perform statistical and probabilistic analysis
• Use error budgets to help your team have better discussions and make better data-driven decisions
• Build supportive tooling and resources required for an SLO-based approach
• Use SLO data to present meaningful reports to leadership and your users

On the surface, this book is about service level objectives (SLOs). But on a deeper level, this book is about people. All the theory, philosophy, and approaches outlined in the pages that follow really only exist to make people’s lives easier, and therefore hopefully better.

We’re going to be discussing a lot of topics. Some of them are going to be fairly philosophical, and some will be heavy with math and formulae. Some will focus on software, while others will focus on processes.

The most important thing that I want you to know when you start reading this book is that this is all only a model. SLO-based approaches to reliability are exactly that—they’re approaches, not some magic tincture that will immediately solve all of your problems. They’re a way to have better discussions and develop measurements that should allow for you to make better data-driven decisions. But there is no one-size-fits-all approach to this. Everything we talk about in this book is only a model.

The book has been split into three parts.

Part I, SLO Development: This should be considered mandatory reading if you’re going to implement an SLO-based approach to reliability—it outlines the concepts, philosophies, and definitions you need to be familiar with to take your first steps. The chapters in this part explain what the various components of SLOs are, how to use them successfully, why they work, and how you can make them work for you.

Part II, SLO Implementation: Here you’ll start to hear from some of my brilliant friends from across the industry. While Part I helps you learn how things work, Part II is much more practical in nature. Together we’ll help you tackle some of the more complicated aspects of implementing an SLO-based approach, including the math you need to have meaningful service level indicators (SLIs) and accurate SLO targets, the most useful alerting for your team, as well as how to get buy-in company-wide.

Part III, SLO Culture: This part discusses how you can make sure that adoption of an SLO-based approach to reliability for your team, organization, or entire company can be engaged with in the most efficient manner. Too many people think of implementing SLOs as “something you do,” in the sense that it’s a singular goal you can check off a list. That isn’t how all of this actually works. SLO-based approaches encourage thinking about the reliability of your services in a different and more accurate way. The chapters in this section ensure that you can implement SLOs in the best possible way, and use the concepts behind them for better communications across teams.

Additionally, the final chapter, Chapter 17, Reliability Reporting, brings everything to a close and contains some of the most useful advice in the entire book. Make sure you don’t skip over that.

To make the most out of the book, I truly hope you read every chapter. The chapters in the first section should be read in order, but the last two sections can be read in whatever order best addresses the needs of your situation. You don’t need to read every chapter consecutively; I hope this book will be as much of a reference for you in the future as it is an educational tool to get you started. It was designed to be both a primer and a source for you to return to again and again.

No matter what your situation is, or what your goals are, I hope at the very least you can walk away from this text remembering the lessons Molly taught us. While this book is mostly about computer systems, the general philosophies are broadly applicable: you can’t be perfect, no one needs you to be perfect anyway, it’s too expensive to try to be perfect, and everyone is happier at the end of the day if you embrace this.

About the Author

Alex Hidalgo is a Site Reliability Engineer and expert at all things related to Service Level Objectives. He developed an interest in computers at a young age, started writing his first BASIC programs at around the age of nine, and remembers the Internet when it was all still text. He eventually turned his hobby into a career, working in various capacities as a network engineer, security engineer, and systems administrator and in many roles within the world of IT support. After moving to New York, he joined Admeld as a Technical Operations Engineer, only to find himself employed by Google a few months later due to acquisition.

At Google, Alex was first introduced to the discipline of Site Reliability Engineering, which connected so closely with him that he wonders how he ever did anything else. Eventually, he found his other calling as an educator, writer, and speaker, traveling all over the world training other Site Reliability Engineers, becoming one of the primary developers of the Coursera Google IT Professional Certification, and contributing to multiple chapters of The Site Reliability Workbook -- most notably “Implementing SLOs” and “SLO Engineering Case Studies.”

Recently, he has joined Squarespace, where his focus is now on spreading the concepts of SLO-based approaches to service reliability -- both internally and across the entire industry. When not sharing his passion for error budgets with others, you can find him scuba diving or watching college basketball. He lives in Park Slope, Brooklyn, with his partner Jen and a rescue dog named Taco. He thinks about SLOs so much he once had a dream about defining some for Taco. Twitter handle: @ahidalgosre