DevOps Dictionary

A straightforward reference for DevOps terms and concepts.

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

W

X

Y

Z

SRE (Site Reliability Engineering)

Reliability

SRE (Site Reliability Engineering) is a way of running software services that treats reliability as an engineering problem. A team sets a measurable target called a service level objective, or SLO (for example, 99.9% of requests succeed), and monitors the service against it. The gap between the target and 100% is the error budget, which tells the team when to ship features and when to work on stability. Repetitive manual work is automated.

Patent-style sketch of Site Reliability Engineering: a service engine is read by a monitoring gauge with an SLO target mark, beside a partly drained tank labelled error budget and a mechanical arm labelled automation turning the crank.

See also Uptime, Chaos Engineering, Circuit Breaker, Dead Letter Queue (DLQ), Error Budget, gRPC Deadline.