liveBook · Manning

Reliability

As explained in Grokking Software Architecture

The degree to which a system performs its intended function consistently over time, including remaining available when required.

Reliability asks whether a system keeps doing the right thing under pressure and when things go wrong. Availability—being up and ready when needed—is related, but reliability also concerns correct behavior when failures occur. section 2.4 section 2.8

A checkout tied to a payment processor

Imagine checkout logic that calls a payment processor directly. If the processor is unavailable, the checkout flow depends on that unavailable service. The tight coupling also makes the flow difficult to unit-test without using the live network. section 10.4.1

Reliability is a quality requirement, not a promise to make every system perfect. Ask which qualities matter to the business and what it can afford. In finance or healthcare, reliability, security, and data integrity may be non-negotiable, even when a deadline is tight. section 2.4 section 2.5.1 section 2.8

How Grokking Software Architecture covers it

How deep it goes
Defines correct behavior under failure and examines a fragile dependency, without detailing a reliability implementation.
Examples
Uses a checkout flow dependent on an external payment processor and contrasts industry constraints in finance and healthcare.
What you can do after
Evaluate reliability risks and priorities when weighing architectural choices against business constraints.
How it connects
Connects reliability with availability, tight coupling, and testability.

Also explained in

Compare how 21 books explain reliability →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Grokking Software Architecture. The section links point to the parts of the book it draws on.