Does RAID replace backups?

No. It does not protect against deletion, ransomware, theft, fire, controller faults or many forms of corruption.

Why is usable capacity lower than expected?

Redundancy, decimal versus binary units, formatting, snapshots and reserved space all reduce what the operating system reports.

Is RAID 5 suitable for large drives?

It can be, but long rebuilds and the risk of another failure make RAID 6 or RAID 10 attractive for some large or important arrays.

Can I add a drive later?

Only if the NAS or controller supports expansion for that RAID type. Check the vendor documentation before buying.

Should all drives be identical?

Matching capacity and performance simplifies planning, although some systems support mixed models.

RAID Capacity Calculator | RAID 0, 1, 5, 6 and 10

SectionTechnology

Last reviewed26 July 2026

Reading time9 minutes

Your dataNothing is saved

What this calculator works out

This calculator gives the usable capacity of an array from the drive count, drive size and RAID level, and rejects combinations that a level cannot support.

Usable capacity is the visible cost of redundancy. The less visible cost is that RAID is not a backup, and treating it as one is the most consequential misunderstanding in this area.

Enter your details

Number of drives

Capacity of each drive (TB)

RAID level

Enter your figures and select Calculate.

Before you choose a level

RAID 0 has no redundancy at all. It combines drives for capacity and speed, and any single failure loses everything.
RAID 1 mirrors, so usable capacity is one drive regardless of how many are in the mirror.
RAID 5 uses one drive's worth for parity; RAID 6 uses two. RAID 10 mirrors then stripes and gives half the raw capacity.
Drives are sold in decimal terabytes and reported in binary, so a 4 TB drive shows as about 3.64 TiB.

RAID protects against drive failure and nothing else

It does not protect against deletion, corruption, ransomware, fire, theft, controller failure or a mistake. All of those affect every drive in the array simultaneously. RAID is an availability measure — it keeps a system running through a drive failure — and a backup is a recovery measure. Needing one does not remove the need for the other.

How this calculator works

Usable capacity by level:

RAID 0: drives × size, minimum 2 drives
RAID 1: size, minimum 2
RAID 5: (drives − 1) × size, minimum 3
RAID 6: (drives − 2) × size, minimum 4
RAID 10: drives ÷ 2 × size, minimum 4 and an even number

The minimums are structural rather than advisory — RAID 5 cannot be built on two drives, and RAID 10 cannot be built on an odd number. The calculator rejects those combinations rather than returning a figure that could not exist.

Worked example: four 4 TB drives

Using the default figures — 4 drives of 4 TB each, RAID 0:

Usable capacity: 16 TB

All sixteen terabytes, and no protection whatsoever — one drive failure loses the lot, and with four drives the chance of that within a few years is considerably higher than with one. The same drives give 12 TB at RAID 5, 8 TB at RAID 6 and 8 TB at RAID 10. The choice is how much capacity to trade for how much resilience, and RAID 0 trades none of it away because it offers none.

Common mistakes

Treating RAID as a backup. It protects against drive failure alone.
Using RAID 0 for anything you care about. It multiplies risk rather than reducing it.
Expecting the decimal capacity to appear. A 4 TB drive reports as about 3.64 TiB.
Mixing drive sizes. Most levels use the smallest drive's capacity for every member.
Filling an array. Performance and rebuild behaviour both degrade near capacity.

Frequently asked questions

Which RAID level should I use?

For a small home NAS, RAID 1 or its equivalent is simple and robust. For four or more drives where capacity matters, RAID 6 or a modern equivalent is generally preferred over RAID 5 because it survives a second failure during a rebuild. RAID 10 gives the best performance and the worst capacity efficiency. RAID 0 is for scratch data you can afford to lose.

Why is RAID 5 discouraged with large drives?

Rebuilding an array after a failure means reading every remaining drive completely, which takes many hours on large drives and stresses them at exactly the moment they are least able to take it. A second failure during that window destroys a RAID 5 array. The larger the drives, the longer the window, which is why RAID 6 has largely displaced it.

Do I still need backups with RAID?

Yes, and this is the central point. RAID does not protect against the things that most often destroy data — deletion, corruption, ransomware, fire, theft. It protects against a drive dying. Three copies of the data, on at least two types of media, with one kept off-site. Each part defeats a different failure: drive failure, accidental deletion, and fire, theft or ransomware.

What about ZFS, Btrfs and Synology SHR?

Modern systems offer equivalents with additional benefits: checksumming that detects silent corruption, snapshots, and in SHR's case the ability to mix drive sizes efficiently. The capacity arithmetic broadly follows the classic levels, but the resilience and flexibility can be substantially better. Worth investigating before defaulting to traditional RAID.

Is what I enter stored?

No. Figures are processed in your browser and never transmitted or retained.

Related tools

References

National Cyber Security Centre — data resilience, backup and ransomware guidance
Information Commissioner's Office — security and availability obligations for personal data

Sources are checked at publication and can change — how I choose and check references.