if businesses need continuous availability they are going to have redundant hardware.
it may be 100% duplicate (1-1 replication), or it may be a n+x style cluster of machines, but in either case they will have the redundant hardware so that they can survive hardware failing.
there are some companies that try to save a little money and make the backup boxes less powerful than the primary boxes, but they almost never do this after they have actually _used_ their backup boxes (people really don't care that you had a system failure and are running on a backup, they just want their stuff to work, and work fast)