I have about 20TB of data that will remain unchanged (forever) and I am planning to make offline copies with a little redundancy.

I already have a lot of 2TB drives that I plan on using and I am thinking of doing 2TB x3 (RAID 5) arrays with a total of 5 sets i.e. 15 drives with 20TB of redundant data.

Couple of things about my data:

  • I can also do 2TBx6 Raid-6 with same number of drives and higher redundancy but since these drives will be offline, I do not want more than 3/4 drives per array (most basic servers have 4 HDDs).

  • My data is reproducible so offsite storage isn't needed and 3-drive RAID-5 is plenty of redundancy.

  • The data will almost never need to be accessed. Max once or twice in a lifetime. ;)

  • The data is mostly mp4 archives with huge 3-4 GB files.

Given my requirements, is my planned methodology sane or do I have a better option?

  • My best advice is just to store all files on several different drives, not to use raid. – Moab Mar 1 at 15:16

For long-term archiving I’d chose raid6 or even consider used entry-level tape library. HDD are not intended for long lifespan, that’s why RAID6 looks more attractive. Here is a good overview of backup media - http://www.hyper-v.io/keep-backups-lets-talk-backup-storage-media/

If I were you, I’d look at cloud storage, especially at archive tiers. AWS Deep Archive and Azure Archive will cost about 20 USD monthly for 20TB of data. If you need to restore once or twice in a lifetime, it makes no sense to maintain local backup infrastructure.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.