RAID Calculator
Enter your disk count, size and RAID level to compare raw and usable capacity, speed characteristics and drive-failure tolerance. Results assume equal-sized disks; RAID does not replace backup.
Result
| Capacity | 48 TB (43.66 TiB) |
|---|---|
| Speed | Fast read / slower write |
| Fault tolerance | 2-drive failure |
| Raw capacity | 72 TB |
| Assumption | Dual distributed parity. |
Want a side-by-side view? Jump to the RAID level comparison below. To estimate rebuild time and unrecoverable read error exposure, use the separate RAID Rebuild Time Calculator.
Compare RAID levels
Swipe the table sideways to see all columns ->
| Level | Min. | Usable | Fault tolerance | Note |
|---|---|---|---|---|
| RAID 0 | 2 | 72 TB | No fault tolerance | Striping only. |
| RAID 1 | 2 | 12 TB | 5-drive failures only with an N-way mirror; a 2-disk mirror tolerates 1 | Modeled as an N-way mirror; most hardware controllers cap RAID 1 at 2 disks and treat extras as spares - use RAID 10 for more than 2 disks. |
| RAID 1E | 3 | 36 TB | At least 1 drive | Striped mirror; controller behavior varies. |
| RAID 5 | 3 | 60 TB | 1-drive failure | Single distributed parity. |
| RAID 5E | 4 | 48 TB | 1-drive failure | Parity plus distributed spare. |
| RAID 5EE | 4 | 48 TB | 1-drive failure | Controller-specific distributed spare. |
| RAID 6 | 4 | 48 TB | 2-drive failure | Dual distributed parity. |
| RAID 10 | 4 | 36 TB | At least 1; one per mirror pair | Even drive count required. |
| RAID 50 | 6 | 48 TB | 1 per RAID 5 group (up to 2 total, if each failure is in a different group) | 2 equal RAID 5 groups striped together, at least 3 disks per group (6 disks gives 3 per group). Disk count must divide evenly by the group count. |
| RAID 60 | 8 | Unavailable | - | 2 equal RAID 6 groups striped together, at least 4 disks per group (8 disks gives 4 per group). Disk count must divide evenly by the group count. |
Note: Capacity assumes equal-sized disks. Filesystem, controller metadata, hot spares, compression, and deduplication are not included. RAID is not backup.
Learn more
How it works
Enter the number of disks, the size of one disk, and the RAID level. The calculator assumes every disk is the same size and works out how much of the raw space holds data and how much goes to parity or mirror copies.
- RAID 0 stripes data with no redundancy: usable space is every disk, and one failure loses the array.
- RAID 1 mirrors: usable space is one disk, however many copies you add.
- RAID 5 gives up one disk of space to parity and survives one failure. RAID 6 gives up two and survives two.
- RAID 10 stripes across mirrored pairs: usable space is half the disks.
- RAID 50 and 60 stripe across several RAID 5 or RAID 6 groups. Each group gives up one (RAID 50) or two (RAID 60) disks, so more groups means less usable space but faster rebuilds and more failures tolerated, as long as they land in different groups.
The dropdown also has RAID 1E, 5E and 5EE, less common layouts that some controllers offer. The Compare RAID levels table has a note on each.
Capacity is shown in decimal TB, the unit drive makers use, and in TiB, the unit most operating systems report. That is why a "4 TB" disk shows up as about 3.64 TiB.
Worked example
Eight 4 TB disks give 32 TB raw. Here is what each common level leaves you:
Swipe the table sideways to see all columns ->
| Level | Formula | Usable | Survives |
|---|---|---|---|
| RAID 5 | (8 - 1) x 4 TB | 28 TB (25.47 TiB) | 1 disk |
| RAID 6 | (8 - 2) x 4 TB | 24 TB (21.83 TiB) | 2 disks |
| RAID 10 | 8 / 2 x 4 TB | 16 TB (14.55 TiB) | At least 1; one per mirror pair |
| RAID 60, 2 groups | (8 - 2 x 2) x 4 TB | 16 TB (14.55 TiB) | 2 per group |
With large disks, RAID 6 is usually the safer pick over RAID 5: a rebuild reads every surviving disk, and a second problem during that window loses a RAID 5 array.
Worked workflow: compare capacity with the rebuild window
Start with six equal 12 TB disks in RAID 6. The capacity calculator shows 72 TB raw and 48 TB usable, with two-drive failure tolerance under its stated model. Hot spares, filesystem and controller overhead are not included.
Now open RAID Rebuild for one 12 TB replacement disk. At 180 MB/s sustained rebuild speed and 30% production load, the model gives 26.5 hours. With a URE exponent of 14, it reports 61.71% per-disk unreadable-sector exposure. That figure is not the chance the array fails and does not model the controller's recovery behavior.
Use the rebuild estimate to ask whether the degraded window is acceptable. Then plan an independent backup and test restoration. RAID tolerance, a rebuild estimate and a backup-capacity estimate answer different questions.
Why is my usable space smaller than the calculator says?
The calculator stops at the array. Filesystem overhead, controller metadata, hot spares, and snapshot reserves all take more space on a real system.
Can I mix disk sizes?
Most controllers treat every disk as the size of the smallest one, so larger disks waste the difference. Enter the smallest disk size for a realistic figure.
How many groups should RAID 50 or 60 use?
The disk count has to divide evenly into groups, with at least 3 disks per RAID 5 group or 4 per RAID 6 group. Twelve disks can be 2 groups of 6, 3 groups of 4, or 4 groups of 3 for RAID 50, for example. More groups trade capacity for faster rebuilds.
Maintained by the RACKCRUNCH team. Last reviewed October 26, 2026. How we calculate | Report a problem