Its mainly the size that creates the issue, but there is another factor.
most 3-4 TB enterprise drives have a non-recoverable read error for every 1 sector per 10^15 bytes.
compared to fx. a 1.2TB 10k 2.5" drive, which has a rate of 10^15.
ive personally had a raid 50 have a non-recoverable read error when when doing a rebuild (on a equallogic array), i was very fortunate, but it could have meant that we would have had to do a full recovery from backups.
With 3par in particular, it makes me wonder a bit more though, since all raid sets are made from small 1GB chuncklets, it should not have the same impact, when doing a rebuild, compared to doing a full rebuild of a physical disk (depending on amount of disks etc.). Raid-MP and Raid-6 might more be a "checkmark" function for 3par, so they can fullfill requirements of raid-6 in offers.
If one of the Guru's could enlighten, it would be appreciated.
|