From the perspective of write amplification and rebuild risk.
Applying the conventional wisdom from the HDD era directly to SSDs is a loss, because rebuild time and write amplification are completely different.
RAID-5 tolerates up to one disk failure. The problem is the load placed on the remaining disks during a rebuild. HDDs take tens of hours to rebuild, so the probability of a second failure during that time is not negligible.
| Method | Minimum disk | Capacity efficiency | Rebuild load |
|---|---|---|---|
| RAID-5 | 3 | N-1 | High (parity recalculation) |
| RAID-6 | 4 | N-2 | Very high |
| RAID-10 | 4 | 50% | Low (mirror copy) |
RAID-10 is safer for workloads with many random writes, such as VM storage. If capacity is important and the workload is mainly sequential writes, a RAID-6 + backup combination is better.
RAID is a means of availability, not a backup. If you place a backup within the same RAID, both the original and the backup are lost together in the event of a controller failure.
If you tell us your site conditions, we will review a configuration that can be applied.
Technical Inquiry →