This just came in the email:
As you are likely aware, this past February one of the two Uninterruptible Power Supply (UPS) systems in the Science Center data center failed completely as the result of a short circuit. . . .
It’s a good thing they had two of them!
But maybe not as good as you think, because redundant components such as the UPS in question don't often fail independently (common design, operating environment, maintenance, etc.), so the joint probability of both failing can be significantly higher than if they failed independently. This is a common issue in analysis of highly reliable systems, which often employ redundancy in design.