The word toil carries a dread for Site Reliability Engineers. It reeks of struggle, drudgery, and incessant, tiring work. Understanding toil, and eradicating it is fundamental to the SRE philosophy.

What is Toil?

The classical definition of toil, as established by Google, is:

“The kind of work tied to running a production service that tends to be manual, repetitive, automatable, tactical, devoid of enduring value, and that scales linearly as a service grows.“

Let’s break down those key characteristics:

• Manual: A human operator is directly involved in the execution, even if it’s just running a pre-written script.
• Repetitive: It’s work you do over and over again, like acknowledging the same recurring alert every morning.
• Automatable: A machine could perform the task just as well as a human, or the need for the task could be designed away. If human judgment is essential, it’s generally not toil.
• Tactical: It’s reactive, interrupt-driven work, rather than proactive, long-term strategic engineering.
• Devoid of Enduring Value: The service remains in the same state after the task is finished. It’s maintenance, not improvement.
• Scales Linearly: As the service grows (more users, more servers, more traffic), the amount of work required also grows proportionally.