10 × 4 TB in RAID 10: the numbers
| Usable space | 20 TB (18.2 TB in Windows) |
|---|---|
| Raw space | 40 TB |
| Used for protection | 20 TB |
| Efficiency | 50% |
| Drives that can fail | 1 to 5 |
| Rebuild at 150 MB/s | 7 h 24 min |
| Rebuild at 50 MB/s (busy) | 22 h |
| Data read during rebuild | 4 TB |
| Read error risk, desktop drives | 27% |
| Read error risk, NAS drives | 3.1% |
Rebuild
Replacing a failed drive means writing a full 4 TB drive. At 150 MB/s with no other load that takes about 7 h 24 min; on a busy array running at 50 MB/s, about 22 h. Only the mirror partner is read, so the other drives are not stressed.
Read errors
Reading 4 TB, the chance of at least one unreadable sector is 27% with desktop drives (1 per 1014 bits) and 3.1% 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
With 10 drives, RAID 6 gives more space (32 TB instead of 20 TB) and survives any two failures. RAID 10 is the better pick for databases and virtual machines that write many small blocks.