RAID Capacity Calculator

Background.

The same eight 4 TB drives can give you 32 TB, 28 TB, 24 TB or 16 TB of usable space, depending only on how they are arranged. Striped as RAID 0 they keep everything and protect nothing. As RAID 5 they give up one drive to parity, as RAID 6 two, and as RAID 10 half the pool goes to mirror copies. What the lost space buys is the number of drives that can die before the data does.

That number comes in two versions, and confusing them is how risky designs get approved. RAID 6 survives any two failures, whichever drives they hit. RAID 10 across eight drives can survive four, but only if each failure falls in a different mirrored pair; two failures in the same pair end the volume. So every result here shows a guaranteed figure and a best-case figure side by side.

Then there is the number on screen. Drive makers count in decimal terabytes of 10¹² bytes. Windows, and many Linux and NAS tools, divide the same bytes by 2⁴⁰ and often still print "TB", so a 24 TB array appears as about 21.83. Nothing is missing; the two figures are one quantity in different units, and both are given.

A few things fall outside this arithmetic. Larger drives are cut down to the size of the smallest member. Hot spares add nothing until they are rebuilt into the array. Filesystem metadata and reserved space take a further slice that varies by filesystem, and rebuild time depends on the drives and controller, so neither is estimated.

What is raid capacity calculator?

RAID capacity is the space a filesystem can use on an array of drives, as opposed to the raw total of the drives fitted; the difference is spent on redundancy. RAID 0 stripes with no redundancy. RAID 1 stores a full copy on every member, so usable space is one drive however many mirrors there are. RAID 5 spreads one drive's worth of parity across all members and RAID 6 two independent parities. RAID 10 stripes across mirrored pairs. RAID 50 and RAID 60 stripe across several RAID 5 or RAID 6 groups, called spans, and pay the parity cost once per span.

How to use this calculator.

1. 01Choose the RAID level. When comparing options, run the same drives through several levels; the capacity and failure figures move in opposite directions.
2. 02Enter the number of member drives, leaving out hot spares. RAID 10 needs an even count.
3. 03Enter the capacity of the smallest member, then say which unit you typed: TB or GB for the figure on the box or invoice, TiB or GiB for a figure copied from an operating system or controller screen.
4. 04For RAID 50 or RAID 60, set the number of spans. The drives must divide evenly between them, with at least three per RAID 5 span and four per RAID 6 span; the field is ignored at other levels.
5. 05Check usable TB against a quotation and usable TiB against what your monitoring will show, and plan around the guaranteed failure count rather than the best case.

The formula.

Everything is worked in bytes first. A drive entered as 4 TB is 4 × 10¹² bytes; entered as 4 TiB it is 4 × 2⁴⁰ = 4,398,046,511,104 bytes.

Each level then leaves u drives' worth of usable space out of N members: u = N for RAID 0, 1 for RAID 1, N − 1 for RAID 5, N − 2 for RAID 6 and N ÷ 2 for RAID 10. The spanned levels follow from their structure. RAID 50 is a stripe over s RAID 5 groups of N ÷ s drives, each keeping all but one, so u = s × (N ÷ s − 1) = N − s; that matches the published rule of N minus the number of logical drives. RAID 60 keeps all but two per span, so u = N − 2s. The stripe on top costs nothing.

Usable bytes are u × C, raw bytes N × C, overhead the difference, and efficiency u ÷ N × 100. Each byte total is then divided by 10¹² for decimal terabytes and by 2⁴⁰ for tebibytes. Their ratio is fixed: one TB is 0.9095 TiB, so TiB figures always read about 9% lower.

The guaranteed failure count is the number of arbitrary drives that can be lost: 0 for RAID 0, N − 1 for RAID 1, 1 for RAID 5, 2 for RAID 6, 1 for RAID 10 and RAID 50, and 2 for RAID 60, because the next loss might land in the pair or span already degraded. The best case assumes every loss lands somewhere different: N ÷ 2 for RAID 10, s for RAID 50 and 2s for RAID 60.

A worked example.

Frequently asked questions.

References& sources.

1. [1]Oracle. Sun StorEdge 3000 Family RAID Basics, Appendix A. Source of the per-level capacity table used here: RAID 0 capacity 'N'; RAID 1 usable capacity 'equivalent to the capacity of one drive'; RAID 5 capacity 'N-1'; RAID 50 capacity 'N - # of logical drives'. Also the source for the minimum and maximum member counts quoted for RAID 0 and RAID 5.
2. [2]Linux kernel. md(4) manual page — the manual for the kernel's own software RAID driver. Source of the fault-tolerance figures ('all but one (N-1) for RAID level 1', 'this number is one for RAID levels 4 and 5', 'two for RAID level 6'), the RAID 6 capacity rule ('it requires N+2 drives to store N drives worth of data'), the two-replica RAID 10 capacity example ('space equivalent to 2.5 of the devices will be available'), and the mixed-drive-size rule that only the space on the smallest device is used.
3. [3]Dell. PowerEdge RAID Controller 9 User's Guide (H330, H730, H830), 'Spanned RAID levels'. Source of the structural definition this page derives the RAID 50 and RAID 60 capacity formulas from: 'a RAID 10 has multiple sets of RAID 1 arrays where each RAID 1 set is considered a span', and 'RAID 50 and RAID 60 combine multiple sets of RAID 5 or RAID 6 respectively with striping'.
4. [4]National Institute of Standards and Technology. Prefixes for binary multiples (Reference on Constants, Units and Uncertainty), citing IEC 60027-2 second edition, 2000-11. Source of Ki = 2¹⁰, Mi = 2²⁰, Gi = 2³⁰, Ti = 2⁴⁰, and of the statement that manufacturers of computer storage devices usually use megabyte to mean 1,000,000 bytes while memory manufacturers use 2²⁰ = 1,048,576 bytes.