Guide
Comparing RAID levels by usable space, redundancy and rebuild time
RAID combines several disks into one volume for capacity, speed or protection. The choice of level decides how much of your purchase you can use and how many failures you survive. The page compares the levels for your disk count and size, so you can see the trade-off before buying drives.
What each level does
RAID 0 stripes data across all disks: full capacity and speed, no protection, and any single failure loses everything. RAID 1 mirrors data, so usable space is one disk’s worth for a two-disk set and failure of all but one is survived. RAID 5 distributes one disk of parity, giving up one disk’s capacity and surviving one failure. RAID 6 uses two parity blocks and survives two. RAID 10 stripes across mirrored pairs, giving half the raw capacity and surviving one failure per pair. JBOD simply concatenates disks and offers neither speed nor protection.
The inputs and results
Choose the level, the number of disks, the disk size in TB and the rebuild speed in MB/s; the default speed is modest because a disk under load rebuilds far slower than its rated speed. Presets load 4 × 4 TB RAID 5, 6 × 8 TB RAID 6 and 8 × 2 TB RAID 10. The result gives usable capacity (also in TiB, which is how many systems report it), raw capacity, efficiency, the number of disk failures survived, the small random write cost and the rebuild time for one disk. A comparison table shows every level for the same disks.
TB versus TiB
Disk vendors use decimal units: 1 TB is 10¹² bytes. Operating systems and storage software often report binary units: 1 TiB is 2⁴⁰ bytes, about 1.0995 × 10¹². A 12 TB array therefore shows as about 10.91 TiB, and no capacity has gone missing. The page shows both so the number on your box and the number on your screen can be reconciled.
A worked example
For four 4 TB disks in RAID 5 the page reports 12.00 TB usable (10.91 TiB), 16.00 TB raw, 75 % efficiency, 1 disk failure survived and a small-write cost of 4 I/Os. At an assumed rebuild speed of 100 MB/s, replacing a disk takes about 11 h 6 min 40 s, because 4,000,000 MB at 100 MB/s is 40,000 seconds. The comparison table shows RAID 0 at 16.00 TB with no protection, RAID 1 at 4.00 TB, RAID 6 at 8.00 TB surviving two disks, and RAID 10 at 8.00 TB.
Limits
RAID is availability, not backup: it protects against disk failure but not against deletion, ransomware, fire, a controller fault or a bad update. The rebuild time is a lower bound that ignores the load on the array and any hot-spare delay. File systems that manage their own redundancy, such as ZFS, are better planned with the ZFS Capacity Calculator, which adds the reserved space and a comfortable fill level.
Related tools
- ZFS Capacity Calculator plans RAIDZ pools including reserved space.
- CCTV Storage Calculator shows how much space video needs before RAID overhead.
- SLA Uptime Calculator turns redundancy into expected availability.
- Byte Converter converts between TB, TiB and other units.
Checked against the current version of the RAID Calculator on .

