RAID Capacity Calculator

Choose RAID 0, 1, 5, 6, or 10.

Enter the total number of installed disks and the labelled capacity of each disk.

Enter any disks that will be reserved as hot spares. Hot spares are removed before the active RAID capacity formula is applied.

The result shows usable capacity, decimal-to-binary capacity, active RAID overhead, hot-spare reserve, storage efficiency, and the basic disk-failure tolerance of the selected level.

Every disk is treated as having the same usable hardware capacity.

The formulas represent ideal RAID-level capacity rather than the exact formatted space reported by a particular controller, operating system, NAS, or filesystem.

RAID 1 is treated as one complete mirror across all active disks. RAID 10 is treated as a stripe across two-disk mirror pairs.

RAID levels trade usable capacity for redundancy and performance characteristics.

The variable n represents active disks after hot spares are removed. S represents the capacity of one equal-size disk.

RAID 0 stripes data across every active disk and uses the entire active capacity for data.

It has no parity or mirror copy. Failure of any active disk makes the striped dataset incomplete.

RAID 0 may be useful for temporary, replaceable, or performance-focused data, but it should not be treated as redundant storage.

RAID 1 mirrors the same data across two or more active disks.

Under the full-mirror model used here, usable capacity equals the size of one disk regardless of how many mirror copies are present.

A two-disk mirror has 50% active-array efficiency. A three-disk full mirror has approximately 33.33% efficiency but retains an additional complete copy.

RAID 5 stripes data and single distributed parity across at least three active disks.

The equivalent capacity of one active disk is consumed by parity, so usable capacity equals (n − 1) × S.

RAID 5 can normally continue after one active disk failure. A second active failure before recovery exceeds the level's parity tolerance.

RAID 6 uses dual distributed parity across at least four active disks.

The equivalent capacity of two active disks is consumed by parity, so usable capacity equals (n − 2) × S.

RAID 6 can normally continue after two active disk failures, although the array may remain degraded until failed disks are replaced and rebuilt.

RAID 10 combines mirroring and striping.

The model used here forms two-disk mirror pairs and stripes data across those pairs, requiring an even number of active disks.

Usable capacity is half the active raw capacity.

RAID 10 always tolerates one disk failure. It may tolerate several failures when they occur in different mirror pairs, but losing both disks from the same pair causes data loss.

A hot spare is an installed disk reserved to replace a failed active disk.

It does not contribute to normal usable capacity, so the calculator subtracts hot spares before applying the RAID formula.

When a redundant active disk fails, a supported controller or software RAID system may rebuild the missing data onto the spare.

A spare can shorten the period before rebuilding begins, but it does not add another parity block or mirror copy to the active layout.

Drive manufacturers normally label capacity using decimal SI prefixes.

One TB means 1,000,000,000,000 bytes.

One TiB means 1,099,511,627,776 bytes, so a decimal terabyte is approximately 0.9095 TiB.

The amount of data is unchanged. The number appears smaller because the binary unit is larger.

This calculator assumes equal-size disks.

Many traditional RAID implementations restrict every member to the usable size of the smallest disk.

For example, a 4 TB disk added to an array of 2 TB disks may contribute only 2 TB to that RAID group, leaving the remaining capacity unused or available only through a separate configuration.

Controller-specific features, hybrid layouts, and storage pools can behave differently, so confirm the actual platform rules before purchasing disks.

The displayed usable value is an ideal RAID-level result.

Actual file storage can be lower after RAID metadata, partition tables, filesystem structures, reserved blocks, snapshots, thin-provisioning reserves, alignment, and platform-specific overhead are applied.

The calculator does not guess these values because they differ across controllers, filesystems, NAS products, operating systems, and configuration choices.

Active-array efficiency compares usable data capacity with the raw capacity of active RAID members.

Installed efficiency compares usable capacity with every installed disk, including hot spares.

A RAID 6 array may have 60% active-array efficiency but only 50% installed efficiency when one additional disk is reserved as a hot spare.

A redundant RAID array normally enters a degraded state after a disk failure.

Rebuilding reconstructs the missing mirror or parity data onto a replacement disk or hot spare.

Rebuild duration and risk depend on disk size, workload, controller behaviour, read errors, remaining-disk health, interface speed, and implementation details.

The capacity calculator does not estimate rebuild time or the probability of another failure.

RAID can keep data available after specific disk failures, but it is not an independent copy of the data.

It does not restore files deleted by a user, overwritten by an application, encrypted by malware, corrupted by software, destroyed with the system, or lost because every copy shares the same failure.

Important data still needs separate, versioned, tested backups stored according to an appropriate recovery plan.

All RAID calculations run in the browser.

Disk sizes, disk counts, RAID selections, and hot-spare settings are not uploaded or permanently stored.