Skip to content
BENIBIX Learn

Site Reliability Engineering (SRE)

Infrastructure📟 ReliabilityService Levels

Also known as SRE, Site Reliability Engineer, Reliability Engineering

Running live systems the way you build code. Set goals, and let code do the dull work.

Think of it like

It is like a pit crew that also builds better cars. They fix things fast, then make them break less.

Example

At Google, SRE teams cap hand-done work, like tickets and on-call duty, at half their time. The rest goes to code that makes the service more reliable.

Why it matters

It turns reliability from a feeling into a number. Teams then argue less and fix the right things.

Definition

Site reliability engineering is a way of running live systems with the tools of software work. An SRE writes code to do the jobs people used to do by hand. The team agrees on how reliable a service must be, and measures it. Time spent on manual work is kept low, so there is time to fix root causes.
SRE work runs on these lines. The team measures the service and spends what is left over.

Common mistake

Giving the old ops team a new name and changing nothing else. With no targets and no time to automate, it is the same job.

Key takeaways

  1. 1
    Measure it. Reliability has a target and a number.
  2. 2
    Automate toil. Repeated hand work becomes code.
  3. 3
    Share the risk. Speed and safety are balanced with a budget.