Single point of failure (SPOF)

A single point of failure is a component with no redundant path around it, so that its failure cascades to everything downstream. The classic examples are a lone identity provider, a non-replicated database, a core switch without a peer, or a cooling unit serving a whole row of racks. Architecture diagrams often claim redundancy that turns out, under load, not to be real.

Single points of failure are best found by running failures rather than by inspection: propagate an outage through a dependency model and measure the blast radius of each asset. Those with the largest reach relative to their size are the SPOFs. Because they concentrate both operational and security risk, removing them tends to be the highest-return investment in a resilience program.