The part that piqued my curiosity...
...is the extensible and variable drive capacities.
I might be a little behind the times, but this is surely a propritary extension to the RAID protocols? As far as I'm aware, all flavours of raid require same-size partitions?
Yes, to can put a larger drive into a standard RAID5, but it will only utilise the same capacity as the other existing drives.
standard RAID provides redundancy by putting staggered checksums on the other drives, so if one fails, the data on it can be recovered from said checksums off the other drives. there's a small but significant storage cost for this, but for the benefit of recovery we (as an industry) have always been happy with that.
For this to work though, the staggering and block sizes have to be standardised across all drives. If you have 4 1TB drives, and a 2TB drive, oboviously, the 2TB will require more checksums to rebuld the data in the event of failiure.
How does this work? Surely You'd have a significant drop in capacity on the 4 smaller drives? The staggering of checksums is also optimised for performance, would non-standard checksum sizes and positions throw this into disarray? (No pun intended)
Enquiring minds would like to know...