Developer & Security

RAID Calculator

Pick a RAID level and your drives to see usable capacity in both TB and TiB, how many drive failures the array survives and how reads and writes compare with a single drive. Mixed drive sizes follow the smallest-drive rule.

Free, runs in your browserUpdated October 2026
RAID level
Drives
TB
Drive size unit
Quick setups

Drive makers use decimal units: 1 TB = 1012 bytes. Windows and many Linux tools count in binary TiB (240 bytes), so the same drive looks about 9% smaller there.

Usable capacity
–

Same drives, every level (tap a row to switch)
LevelUsableEfficiencySurvives
RAID Calculator diagram: four 4 TB drives in RAID 5 give 12 TB usable, shown as 10.91 TiB
How the RAID Calculator works: Usable space, failures survived and TB versus TiB for any RAID level

How to Use the RAID Calculator

How to use the RAID Calculator: choose a RAID level, set drive count and size, read usable capacity and tolerance
Numbered steps on the RAID Calculator. Follow them in order.
  1. Pick the RAID level: 0, 1, 5, 6 or 10.
  2. Choose identical drives, or mixed sizes to apply the smallest-drive rule.
  3. Enter how many drives you have and the size printed on each one.
  4. Read the usable capacity in TB and TiB, then compare every level in the table below.

Choose a RAID level, then describe your drives. With Identical drives, enter how many drives you have and the size printed on each one. With Mixed sizes, list every drive, such as 4, 4, 6, 8. Pick TB or GB to match the label on the drives. The panel shows the usable capacity in decimal TB and in binary TiB, a bar that splits the raw space into usable, redundancy and unused parts, the number of drive failures the array can survive, and theoretical read and write speed compared with a single drive.

The table at the bottom of the result panel repeats the calculation for every level with the same drives, so you can compare RAID 5 with RAID 6 or RAID 10 at a glance. Tap a row to switch to that level.

RAID Capacity Formulas

With n drives and S as the size of the smallest drive, usable capacity is:

RAID 0 = n × S    RAID 1 = S
RAID 5 = (n − 1) × S    RAID 6 = (n − 2) × S
RAID 10 = (n ÷ 2) × S
efficiency = usable ÷ raw capacity
LevelMinimum drivesSurvivesSmall write costBest for
RAID 02No failures1 I/OScratch and cache space
RAID 12All but one drive1 I/O per copyBoot drives, small servers
RAID 531 drive4 I/OsRead-heavy file storage
RAID 64Any 2 drives6 I/OsLarge arrays, big drives
RAID 104, even1 per mirror pair2 I/OsDatabases, virtual machines

The write cost is the classic write penalty: a small random write on RAID 5 reads the old data and parity, then writes new data and parity. Sequential writes on parity arrays avoid most of it.

Worked Example

Four 4 TB drives in RAID 5 give (4 − 1) × 4 TB = 12 TB usable out of 16 TB raw, a 75% efficiency, and the array survives one failed drive. In binary units that is 12 × 1012 ÷ 240 = 10.91 TiB, which Windows shows as 10.9 TB. The same drives in RAID 6 or RAID 10 give 8 TB, survive two failures in RAID 6, and survive one guaranteed failure in RAID 10.

With mixed drives of 4, 4, 6 and 8 TB in RAID 5, the smallest drive sets the size of every member: (4 − 1) × 4 TB = 12 TB usable from 22 TB raw. The extra 2 TB and 4 TB on the larger drives, 6 TB in total, stay unused by the array.

TB Versus TiB

Drive makers count in decimal units, so 1 TB is 1,000,000,000,000 bytes. Many operating systems count in binary units, where 1 TiB is 240 = 1,099,511,627,776 bytes, and Windows labels that unit TB. The gap is about 9% at the terabyte scale, which is why a new array always looks smaller than expected. The NIST page on binary prefixes explains the official names. File system metadata and reserved space reduce the final free space a little further.

Limits and Practical Advice

  • RAID is not a backup. It keeps a server running through a drive failure, but deleted files, ransomware and controller faults affect every copy. Keep separate backups.
  • Speed figures are theoretical. Real results depend on the controller, caching, file system and workload.
  • Rebuilds take time. A parity rebuild reads every surviving drive. On large drives this can take many hours, so RAID 6 or RAID 10 is safer than RAID 5 for big arrays.
  • Hot spares sit outside the array, so do not count them as drives here.
  • Systems such as ZFS, Btrfs and some NAS hybrid modes can use mixed sizes more efficiently than classic RAID. This calculator follows the standard smallest-drive rule.

For network storage planning, the SNIA dictionary defines each RAID term used here.

Frequently asked questions

How do you calculate RAID 5 capacity?

Multiply the number of drives minus one by the size of the smallest drive. Four 4 TB drives give 3 times 4 TB, which is 12 TB usable. One drive's worth of space holds the distributed parity.

Is RAID 6 better than RAID 5?

RAID 6 survives any two drive failures, while RAID 5 survives only one. It costs one more drive of capacity and slower small writes, but it is safer for large drives and arrays with long rebuild times.

Why does my RAID show less space than the calculator?

Windows and many tools show binary TiB but label it TB, so 12 TB appears as about 10.9 TB. File system metadata and reserved space take a little more.

Can I mix drive sizes in a RAID array?

Yes, but classic RAID treats every member as the size of the smallest drive. The extra space on larger drives is unused until you replace the smaller drives and expand the array.

How many drives can fail in RAID 10?

At least one. RAID 10 survives more failures only when each failed drive is in a different mirror pair. Losing both drives of the same pair loses the whole array.

What is the difference between TB and TiB?

A TB is 10 to the power 12 bytes, used on drive labels. A TiB is 2 to the power 40 bytes, about 1.0995 TB, and is what many operating systems actually measure.