3 ms·
I don't think that is the main problem with that calculation. They assume to know the risk of failure over one year. Having three separate copies would allow yo
by sengstrom 13y ago
I don't think that is the main problem with that calculation. They assume to know the risk of failure over one year. Having three separate copies would allow you to estimate the risk as they do if they don't intend to monitor failures over shorter times than that (which would be a very very bad idea). Therefore the risk is actually much smaller since one would presume that they do something about it when the first hard drive fails...
As noted elsewhere this is hardly the only source of potential data loss however.