Site Reliability Engineering (SRE)
Applying software engineering principles to operations problems.
"SRE is what happens when you ask a software engineer to design an operations team." — Ben Treynor Sloss, Google
Core Concepts
- SLAs, SLOs, and SLIs:Defining reliability targets. Service Level Indicators (metrics), Objectives (goals), and Agreements (contracts).
- Error Budgets:The allowable threshold for failure. If you burn your budget, innovation stops to focus on stability.
- Toil Reduction:Automating repetitive, manual work. SREs should spend 50% of time on engineering, not tickets.