Redundancy

'Redundancy' means to duplicate a piece of hardware (or software) within a system so that if one part fails, the others take over.

For example an extremely expensive satellite would be built with triple-redundant power supplies. If one, fails then the others sense it and switch off the faulty one.

Another example are the RAID disks commonly used in file servers. This is an arrangement where a number of hard disks record the same data. If one fails, then the fault is detected and the other hard disks just carry on whilst technicians replace the faulty one.

Redundancy is also important in safety critical software and works by a 'voting' system. Three pieces of code, written by different developers each are given exactly the same input. Each is expected to produce exactly the same output. If a software fault (bug) turns up in one, it is very unlikely that the same fault will turn up in the other two versions as they were written by a different team - so the three outputs are compared and vote is taken on the correct outcome. Places like nuclear power plants use this method.

Challenge see if you can find out one extra fact on this topic that we haven't already told you

Click on this link: Data Redundancy

2020-10