PPP - indeed there is a lot about defragging, this is why I used to do it near obsessively, however it may be the placebo effect -- I did do a search for actual benchmarks before and after defragmenting drives with particular usage patterns and found nothing.
It will make a difference on some level but I have yet to see proof (versus instructions on how to defrag / adverts for defrag programs) that this difference is anything but negligible.
Can anyone post links to benchmarks?
PS ZH785 -- a failed defrag operation shouldn't lose any data; the data is copied from its source to destination, then the filesystem updated to point to the new block(s) instead of the old.