0

Source-:https://cs.ccsu.edu/~stan/classes/CS410/Notes16/11-ReliabilityEngineering.html

This is self monitoring architecture. So here computations carried across 2 channels, if they both provide same result then system is operating correctly else failure has occurred. Hardware and software in each module is different to avoid common mode failure.

This is N version programming. Here also software should be written by different person, different algorithm, different programming language etc. So what's the difference, both seem same thing to me enter image description here

enter image description here

cuajiu
  • 9
  • 1

1 Answers1

3

The difference is in what is done if the outputs are different:

  • In the self-monitoring architecture, if the outputs are different then a fault is indicated; no recovery is possible - i.e. this is a fault detection architecture.
  • In the N-version architecture, the output selector is a voting mechanism which selects the output which is agreed on by the majority - i.e. it allows for both fault detection and limited recovery, The limitation is that it can recover from a failure in strictly less than half the implementations, which is why it should always be used with N odd.
Philip Kendall
  • 22,899
  • 9
  • 58
  • 61