What is Site Reliability Engineering (SRE)? How does it relate to DevOps?

Site Reliability Engineering (SRE) is Google's discipline for running large-scale systems by applying software engineering practices to operations.

• Origin — coined by Ben Treynor at Google (2003). Predates 'DevOps' as a term.

• Core concepts:

  • SLIs / SLOs / Error Budgets — quantitative reliability targets and how much downtime you can 'spend'.

  • Toil reduction — at most 50% of an SRE's time on toil; the rest on engineering automation.

  • Blameless postmortems — learn from incidents, don't punish.

  • Embracing risk — 100% reliability is the wrong goal; aim for the SLO.

SRE vs DevOps:

• DevOps = cultural philosophy for delivery.

• SRE = concrete implementation of that philosophy with measurable engineering practices.

• Often paraphrased: 'class SRE implements DevOps'.