Site Reliability Engineering: Measuring and Managing Reliability

Overview

Service level indicators (SLIs) and service level objectives (SLOs) are fundamental tools for measuring and managing reliability. In this course, students learn approaches for devising appropriate SLIs and SLOs and managing reliability through the use of an error budget.

Syllabus

Introduction to SRE

This module is intended to bring you up to speed on the concepts underpinning SRE, CRE, and SLOs. If you're already familiar with these concepts, you may still find new information and perspectives in this module, but it is not necessary to complete it.

Targeting Reliability

In this module we’re going to talk about how you measure the desired reliability of a service. We will address what to consider when setting SLOs for your application within your organization. We'll look at the three principles we use to measure the desired reliability of a service: figuring out what you want to promise and to whom, figuring out the metrics you care about that make your service reliability “good", and finally, deciding how much reliability is good enough.

Operating for Reliability

In this module, we’ll start by introducing a mechanism for quantifying unreliability using something called an error budget. We'll show how error budgets help you decide when to focus on making a service more reliable. And then we'll learn about some of the engineering and operational improvements that can help you do that.

Choosing a Good SLI

In this module we will start off by taking a look at some characteristics of monitoring metrics that can make them useful as SLIs and contrast these against other metrics that are less useful. Because the choice of where to measure an SLI is a key variable, we'll cover the five main ways you can measure an SLI and compare their pros and cons.

Developing SLOs and SLIs

In this module, we'll start off with an overview of our four step process for developing SLOs and SLIs for a user journey. We'll introduce the fictional company that created our example mobile game, the infrastructure that we'll be working with, and the simple user journey we'll be applying the four step process to.

Quantifying Risks to SLOs

In this module we'll be taking a critical look at the availability risks for our example service. We want to answer the question: "are our SLO targets and error budgets realistic?"

Consequences of SLO Misses

In this module, we'll cover best practices for documenting your SLOs, the rationale behind a formal error budget policy and how best to create one and finally, we'll look at an example error budget policy in order to understand the trade-offs and incentives that play out during negotiations when trying to write an error budget policy.