2 × 10 TB in RAID 1: the numbers
| Usable space | 10 TB (9.1 TB in Windows) |
|---|---|
| Raw space | 20 TB |
| Used for protection | 10 TB |
| Efficiency | 50% |
| Drives that can fail | Any 1 |
| Rebuild at 150 MB/s | 18 h |
| Rebuild at 50 MB/s (busy) | 2.3 days |
| Data read during rebuild | 10 TB |
| Read error risk, desktop drives | 55% |
| Read error risk, NAS drives | 7.7% |
Rebuild
Replacing a failed drive means writing a full 10 TB drive. At 150 MB/s with no other load that takes about 18 h; on a busy array running at 50 MB/s, about 2.3 days. Only the mirror partner is read, so the other drives are not stressed.
Read errors
Reading 10 TB, the chance of at least one unreadable sector is 55% with desktop drives (1 per 1014 bits) and 7.7% with NAS drives (1 per 1015). Nothing is left to correct it, so the affected file, or on some controllers the whole rebuild, fails. These are worst-case datasheet rates; real drives usually do better.
Our take
A mirror is the simplest safe setup for two drives, and either drive can usually be read on its own. If you add drives later, three 10 TB drives in RAID 5 would give 20 TB, and four in RAID 6 20 TB with two-drive protection.