Ketan Gangatirkar, Indeed, Inc.
The goal of a Site Reliability Engineer is to create a reliable, scalable, performant service. To drive the right results, you need the right measures. Enter Service Level Objectives.
A good SLO starts with your customers. Next understand the role your service plays in your business. Then turn the understanding into numbers. Dive into the data to discover where diminished reliability and performance harm user experience and business results. Only then you can set good objectives.
Good SLOs will mirror your users' and business needs. When your users hurt, you want to know it. By feeling your users' pain you can discover the levels of service that hit the sweet spot of reliability, product success, and operability. It's empathy, but measured and quantified.
This talk will walk you step by step through finding, gathering, and understanding the inputs that inform good SLOs. We will review different classes of SLO and when to use them, with real examples to guide you. When you leave this talk, you'll be armed with practical advice to measure, validate, and improve the reliability of your products.
Ketan Gangatirkar, Indeed, Inc.
Ketan Gangatirkar is the Vice President of Engineering for Indeed's Job Seeker products. For the last 9 years, he's been helping millions of people get jobs. He has broken Indeed's site in dozens of different and creative ways over the years and has finally learned what not to do. For a time, he was responsible for the SRE organization at Indeed, helping the company evolve from centralized operations to a faster, more independent, and more scalable model, so that people like Ketan can't break the site anymore.
Open Access Media
USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.
author = {Ketan Gangatirkar},
title = {Quantifying Empathy with Service Level Objectives},
year = {2018},
publisher = {USENIX Association},
month = jun
}