What is a RAID Capacity Calculator?
RAID (Redundant Array of Independent Disks) combines multiple physical drives into a single logical volume for better performance, redundancy, or both. Because parity and mirroring consume some of your raw storage, the usable capacity is always less than the sum of the disks (except RAID 0). This calculator instantly shows the usable capacity, the space lost to redundancy, and the storage efficiency for RAID 0, 1, 5, 6, and 10.
How to use it
Pick your RAID level, enter the number of disks in the array, and the capacity of each disk (assuming all disks are the same size). The calculator returns usable terabytes, raw total, overhead, fault tolerance, and efficiency percentage.
The formulas explained
For an array of n equal disks of size d: RAID 0 uses everything, so usable = \(n \cdot d\) (no fault tolerance). RAID 1 mirrors all disks, leaving usable = \(d\). RAID 5 reserves one disk of parity: usable = \((n-1) \cdot d\). RAID 6 uses double parity: usable = \((n-2) \cdot d\). RAID 10 stripes across mirrored pairs: usable = \(\left\lfloor n/2 \right\rfloor \cdot d\).
Worked example
Suppose you have 4 disks of 4 TB each in RAID 5. Usable:
$$\text{Usable} = (4-1) \times 4 = 12\ \text{TB}$$Raw total is 16 TB, so 4 TB (one disk) is lost to parity, giving 75% efficiency and tolerance for one drive failure.
RAID Level Comparison for the Same Disk Set
The tables below compare each level for the same set of identical 4 TB disks (decimal TB). Overhead is raw capacity minus usable capacity. RAID 1 is shown as an n-way mirror (every disk holds a copy), which is how the usable-capacity formula treats it. Many controllers instead limit RAID 1 to two disks, so check your hardware.
4 × 4 TB (16 TB raw)
| Level | Min. disks | Usable (TB) | Overhead (TB) | Efficiency | Guaranteed fault tolerance |
|---|---|---|---|---|---|
| RAID 0 | 2 | 16 | 0 | 100% | 0 disks |
| RAID 1 | 2 | 4 | 12 | 25% | n − 1 = 3 disks |
| RAID 5 | 3 | 12 | 4 | 75% | 1 disk |
| RAID 6 | 4 | 8 | 8 | 50% | 2 disks |
| RAID 10 | 4 (even number) | 8 | 8 | 50% | 1 guaranteed; up to 2 if each mirrored pair keeps one disk |
8 × 4 TB (32 TB raw)
| Level | Min. disks | Usable (TB) | Overhead (TB) | Efficiency | Guaranteed fault tolerance |
|---|---|---|---|---|---|
| RAID 0 | 2 | 32 | 0 | 100% | 0 disks |
| RAID 1 | 2 | 4 | 28 | 12.5% | n − 1 = 7 disks |
| RAID 5 | 3 | 28 | 4 | 87.5% | 1 disk |
| RAID 6 | 4 | 24 | 8 | 75% | 2 disks |
| RAID 10 | 4 (even number) | 16 | 16 | 50% | 1 guaranteed; up to 4 if each mirrored pair keeps one disk |
Adding disks improves the efficiency of RAID 5 and 6 (the parity cost is fixed at one or two disks), while RAID 10 stays at 50%. RAID 5 and 6 also have longer rebuild times on large disks, which raises the risk of a further failure during a rebuild. RAID is not a backup.
RAID Level Reference: Minimum Disks and Efficiency
Here n is the number of disks and s is the capacity of the smallest disk (arrays normally use the smallest member's size for every disk). Performance notes are general; real results depend on controller, workload and disk type.
| Level | Minimum disks | Usable capacity | Efficiency | Performance (general) |
|---|---|---|---|---|
| RAID 0 | 2 | n × s | 100% | Fastest reads and writes (striping); no redundancy |
| RAID 1 | 2 | s (all disks mirror each other) | 1/n (50% with 2 disks) | Good reads; writes about as fast as a single disk |
| RAID 5 | 3 | (n − 1) × s | (n − 1)/n | Good reads; writes slowed by parity calculation |
| RAID 6 | 4 | (n − 2) × s | (n − 2)/n | Good reads; heavier write penalty than RAID 5 (double parity) |
| RAID 10 | 4, and an even number | (n / 2) × s | 50% | Good reads and writes; fast rebuilds by copying a mirror |
The calculator's general formula for parity levels is \(\text{Usable} = (n - p) \times s\), where p = 1 for RAID 5 and p = 2 for RAID 6. Usable space shown by an operating system will be lower because of filesystem overhead and the TB/TiB difference (see below).
TB to TiB Conversion for Usable Capacity
Disk makers label capacity in decimal units (1 TB = 1012 bytes), while many operating systems report binary units (1 TiB = 240 bytes = 1,099,511,627,776 bytes). The conversion factor is:
$$\frac{10^{12}}{2^{40}} \approx 0.9095$$
| Usable (TB) | Usable (TiB, approx.) |
|---|---|
| 1 | 0.9095 |
| 4 | 3.638 |
| 8 | 7.276 |
| 12 | 10.914 |
| 16 | 14.552 |
| 24 | 21.828 |
For smaller units, 1 GB = 109 bytes and 1 GiB = 230 bytes, so 1 GB ≈ 0.9313 GiB.
Example: a RAID 5 array of 8 × 4 TB disks has 28 TB usable, which is \(28 \times 0.9095 \approx 25.47\) TiB before filesystem overhead.
FAQ
Why is my usable capacity lower than the marketing number? Drive makers use decimal terabytes (\(10^{12}\)) while operating systems report binary tebibytes (\(2^{40}\)), about 9% smaller. This tool works in whatever unit you enter; format the result accordingly.
Does RAID replace backups? No. RAID protects against drive failure, not accidental deletion, ransomware, or controller faults. Always keep separate backups.
What if my disks are different sizes? Most arrays size each disk to the smallest member, so wasted space occurs. For accurate results, use the smallest disk's capacity as the per-disk value.