About the RAID Calculator
This RAID calculator shows how much storage you actually get from an array of identical drives, how many of them can fail before you lose data, and how the RAID level affects random read and write performance. Pick a RAID level, the number of drives and the capacity of each drive, and it returns usable capacity, the space spent on parity or mirroring, storage efficiency and the write penalty.
It is for home-lab builders sizing a NAS (Synology, QNAP, TrueNAS, Unraid-style setups), sysadmins planning server storage and anyone comparing RAID 5 against RAID 6 or RAID 10 before buying disks. It also converts the drive maker’s decimal terabytes into the binary tebibytes (TiB) that Windows, macOS and Linux report, which explains why a "4 TB" drive shows as about 3.64 TB.
The math assumes all drives are the same size — with mixed sizes, standard RAID treats every drive as the smallest one. Capacities are shown in decimal terabytes (GB inputs are converted at 1,000 GB = 1 TB) and are before file-system overhead (typically a few percent). Remember that RAID protects against drive failure, not against deletion, ransomware or a failed controller, so it is not a backup.
With the default inputs, the usable capacity is 24 TB. Change any value above to recalculate instantly.
How to use the raid calculator
- 1Choose the RAID level you are considering.
- 2Enter how many drives the array will have and the capacity printed on each drive.
- 3For RAID 50 or 60, set how many parity groups the drives are split into.
- 4Optionally enter per-drive IOPS to estimate random performance.
- 5Compare usable space and failure tolerance across levels in the table.
Formula and method
Usable capacity is the number of data drives multiplied by the size of one drive. RAID 0 stripes across every drive with no redundancy; RAID 1 mirrors the same data onto every drive; RAID 5 spends one drive’s worth of space on distributed parity and RAID 6 spends two; RAID 10 mirrors pairs and stripes across them, halving capacity; RAID 50 and 60 stripe across G RAID 5 or RAID 6 groups, each losing one or two drives to parity.
Theoretical random IOPS use the standard write-penalty model: read IOPS ≈ N × per-drive IOPS, and write IOPS ≈ N × per-drive IOPS ÷ penalty, where the penalty is 1 for RAID 0, 2 for RAID 10 and two-way RAID 1 (N for an N-way RAID 1 mirror), 4 for RAID 5/50 (read data, read parity, write data, write parity) and 6 for RAID 6/60. Controller caches and SSDs change real-world numbers. TiB = bytes ÷ 2⁴⁰.
- N
- Number of drives
- S
- Capacity of one drive (smallest drive if sizes differ)
- G
- Number of parity groups (spans) for RAID 50/60
Worked examples
Four 8 TB drives in RAID 5
RAID 5 uses one drive’s worth of space for parity, so 4 × 8 TB gives (4 − 1) × 8 = 24 TB usable out of 32 TB raw — 75% efficiency. 24 TB is 24 × 10¹² ÷ 2⁴⁰ ≈ 21.83 TiB in the operating system. With 150 IOPS per drive, reads reach about 600 IOPS and writes about 600 ÷ 4 = 150.
Six 4 TB drives in RAID 10
RAID 10 mirrors drives in pairs and stripes across the three pairs, so usable space is 6 ÷ 2 × 4 = 12 TB (50%). Any single drive can fail; up to three can fail as long as no two are in the same mirror. Writes cost two disk I/Os, so 6 × 150 ÷ 2 = 450 write IOPS.
Twelve 16 TB drives in RAID 60 with two groups
Two RAID 6 groups of six drives each lose two drives to parity, leaving 12 − 2 × 2 = 8 data drives: 8 × 16 = 128 TB usable of 192 TB raw (66.7%). Each group survives two failures, so the array survives any two and up to four if they are split across groups.
Frequently asked questions
How much usable space does RAID 5 give?+
RAID 5 gives the capacity of all drives minus one: (N − 1) × drive size. Four 4 TB drives give 12 TB, five give 16 TB. The larger the array, the higher the efficiency, but only one drive may fail at a time.
RAID 5 vs RAID 6 vs RAID 10 — which should I choose?+
RAID 5 maximises space but survives only one failure, which is risky with large drives. RAID 6 survives any two failures at the cost of a second parity drive. RAID 10 gives the best write performance and fast rebuilds but only 50% capacity.
Why does my 4 TB drive show as 3.64 TB?+
Drive makers use decimal units (1 TB = 10¹² bytes) while operating systems count in binary units (1 TiB = 2⁴⁰ bytes ≈ 1.0995 × 10¹² bytes). 4 × 10¹² ÷ 2⁴⁰ ≈ 3.64, so nothing is missing.
Can I mix drive sizes in a RAID array?+
Standard RAID uses only as much space on each drive as the smallest member offers, so a 4 TB drive in an array of 2 TB drives wastes 2 TB. Some NAS systems (for example Synology SHR or ZFS mirrors of different vdevs) handle mixed sizes more efficiently.
Is RAID a backup?+
No. RAID keeps data available when a drive fails, but deletions, file corruption, ransomware, fire or a failed controller affect the whole array at once. Keep separate backups, ideally following the 3-2-1 rule.
What is the RAID write penalty?+
It is the number of disk operations one random write costs. RAID 0 needs 1, RAID 10 and two-drive mirrors need 2 (an N-way RAID 1 mirror needs N), RAID 5 needs 4 (read data and parity, write both) and RAID 6 needs 6 because two parity blocks are updated.