I have seen a lot of horror stories, and that is just one of them. A lot of consideration needs to be given to the whole setting when designing these things; it's not just a case of dropping in clever equipment and sitting back.
There are a lot of ways of incorporating RUPS systems into a data centre, but without good discrimination between levels of protection and fault control, bad things can happen. The AWS outage is a classic case of poor design of protection settings.
Some data centres have better redundancy than others, but all are vulnerable if the design is not carefully considered.
The Piller design is the best I have seen, a hell of a lot has to go wrong to have a major power outage with the IP bus system in operation. Even with a major disaster internally, the system isolates the fault extremely quickly with no effect on other parts of the installation.
The first time I saw it was about 5 years ago, and I still admire the depth of engineering that went into the concept.
NXT Price at posting:
$14.05 Sentiment: Buy Disclosure: Held
A personalised tool to help users track selected stocks. Delivering real-time notifications on price updates, announcements, and performance stats on each to help make informed investment decisions.
(20min delay)