WEBINAR ON-DEMAND
High reliability systems: Dual modular redundancy, or very high software integrity levels?

Many engineering principles apply equally to hardware or software engineering. But the very nature of software is that it is pure “thought-stuff”; there is no possibility of an apparently random failure in the same way that there might be for a component failure demonstrating the vagaries of a normally distributed life expectancy. All software failure is systemic by definition. 

This presentation discusses why, as a consequence, the adoption of dual modular redundancy (DMR) as a means for achieving high reliability in devices deploying any form of control logic needs to be approached with caution. It considers alternative methods of addressing that conundrum by means of dual development paths, and contrasts them with the additional development overhead implied in improving the reliability of a single module. And it shows how an integrated tool suite can optimize the effort required to achieve such an improvement.

Login or register to view