Why Do Computers Stop and What Can Be Done About It?

Jim Gray

Analyses real outage data to argue that software bugs and operator mistakes, not hardware, are the dominant cause of system failure, reframing fault tolerance as a software and process problem rather than a hardware one.

Read it

Where it is now wrong

The failure statistics are measurements of Tandem systems in the early 1980s, and the hardware numbers should be read as history rather than as current base rates. The taxonomy they support, that most outages are software and operations rather than hardware, has aged far better than the figures.

Before you start

FreeIntermediatelink checked 17 Sept 2026

Filed under Distributed Systems in Computer Science.