I'm currently using 3 RAID5 controllers in each of several media systems. This is being done in an attempt to preempt future data loss. I store large amounts of scanned, video and audio historical data that can't re replaced so I want to make sure that the loss of a single drive or even two in one array will not mean the permanent loss of all data. Recently we had 2, 3TB 3.5" non-RAID drives fail after less than 2 years of intermittent use. As a result we lost over 2TB of non-replaceable data. Please don't harp about backup, storing that much data in the "cloud" or with some company like Carbonite is cost prohibitive. We expect to have over 40TB of data on similar RAID5 storage systems by the end of the year. This data is the written, audio, video and photographic memories of thousands of WWII vets. We are using RAID5 clusters so as to minimize the effect of loosing a drive.
These systems are currently set up with 3 RAID5 controllers each (HighPoint, RocketRaid), a set of 4 (1.5TB) drives on each. We are using 2.5" NON-enterprise drives because they were donated to us. The system is used anywhere from 0.5 to 6 hours a day. This is not a 24/7 server operation. We have many available drives of the same type to swap when needed. Our goal is to reach 100TB of "lossless" storage by years end.
The question is: Using these drives in this manner do I need to expect a higher than normal failure rate ? Or is the failure rate the same as normal home use ? Or is there a better way to do this using the available 1.5TB drives we have ? (2 Gross)
I thank you in advance for any help you may offer.