Keep systems reliable, without slowing down how fast you ship.
The engineering practices — SLOs, error budgets, incident response — that balance reliability against velocity.
What is site reliability engineering?
Site reliability engineering (SRE) applies software engineering practices to operations problems, with the goal of keeping systems reliable without slowing down how fast teams can ship. At its core are measurable reliability targets — SLOs and SLIs — and the error budgets that translate those targets into a shared, data-driven way to balance new releases against stability.
This pillar covers the practices that make that balance work in practice: incident management, chaos engineering, resilience design, and the blameless postmortems that turn every incident into a lasting fix.
Explore Site Reliability Engineering.
A closer look at the topics that make up this pillar. Each one links out to a dedicated guide.
SLOs / SLIs
Defining what reliability means for a service, in measurable terms.
TopicError Budgets
The allowed threshold of failure that balances reliability against release velocity.
TopicReliability
Engineering practices that keep systems available and correct over time.
TopicIncident Management
Detecting, responding to, and resolving production incidents.
TopicChaos Engineering
Deliberately injecting failure to test system resilience before it’s tested for you.
TopicResilience
Designing systems that degrade gracefully instead of failing completely.
TopicPostmortems
Turning incidents into lasting fixes through blameless review.
Balance reliability with velocity.
Explore the SRE practices that keep systems dependable without slowing down delivery.
