redundancy

Having more than one of a component, such as a server, power feed or network link, so the service keeps running if one of them fails.

In more detail

A redundant pair fails only when both copies fail at once, which is far less likely than one failing, provided the copies fail independently. Shared dependencies break that assumption: two servers on the same power strip, in the same building or running the same faulty update can fail together. It is the main tool of high availability, and it is proven only by testing that the spare really takes over.

Example

def pair(a): return 1 - (1 - a) ** 2   # either of two copies will do

Related terms

Taught in

References