What is the difference between RAID 5 and RAID 6?

RAID 5 reserves one disk’s worth of distributed parity and tolerates any one disk failure. RAID 6 reserves two and tolerates any two. Four equal 4 TB disks therefore give 12 TB in RAID 5 or 8 TB in RAID 6 before filesystem and controller overhead. These are theoretical failure counts assuming healthy remaining disks and a functioning implementation. Unreadable sectors, rebuild duration and the probability of data loss are not estimated.

How much usable space does RAID 10 give?

This conventional RAID 10 model mirrors disks in pairs and stripes across the pairs, so eight equal 4 TB disks give 16 TB usable from 32 TB raw. It requires an even disk count of at least four. Any single disk failure is tolerated; up to four are possible only if each is in a different pair. Two failures in the same pair lose that pair. No transfer-speed or rebuild-time estimate is supplied.

Why does a 12 TB array show up as 10.9 TB?

Decimal TB and binary TiB count the same bytes with different divisors. One TB is 1,000,000,000,000 bytes; one TiB is 1,099,511,627,776. Thus 12 TB is about 10.91 TiB. Some software uses binary units but displays a TB label, while other software uses decimal units. The switch changes this display conversion only. Actual formatted capacity also depends on filesystem and controller overhead.

What happens if the disks are different sizes?

The conventional striped/mirrored calculations here use one common size, usually the smallest disk: a 6 TB disk beside 4 TB disks contributes 4 TB to that model. Unequal mirror pairs and vendor-specific mixed-disk pools may differ. Independent JBOD capacity is the sum of the disks, but this tool accepts one size; add mixed sizes separately. A spanned JBOD volume may lose access to more than the failed disk’s data.

What are RAID 50 and RAID 60?

RAID 50 stripes across equal RAID 5 groups and reserves one disk per group. RAID 60 uses equal RAID 6 groups and reserves two per group. Twelve 4 TB disks in three groups give 36 TB in RAID 50 or 24 TB in RAID 60. RAID 50 tolerates any one failure, or up to three if each is in a different group. RAID 60 tolerates any two, or up to six if no group loses more than two. This tool enumerates valid equal group counts.

No. Redundancy can keep an array available through specified disk failures, but it does not provide a separate historical copy of data. Deletion, corruption, ransomware and shared hardware failures can affect the array. This calculator compares capacity and theoretical disk-failure tolerance; it does not plan or verify backups.

Does this calculate RAID speed, IOPS or rebuild time?

No. The comparison explains how data, mirrors and parity are arranged, but supplies no throughput, IOPS, rebuild duration or unreadable-sector probability. Those require details about disks, controller, workload and software.

Does this reproduce Synology SHR, ZFS RAIDZ or a NAS vendor’s capacity calculator?

No. This is a conventional equal-size RAID model. SHR, RAIDZ, Btrfs, vSAN and proprietary controller or NAS layouts may have different allocation, metadata, reserve and failure behavior. Disk size is entered in decimal GB or TB; the result can be displayed in TB or TiB.

Are hot spares included in the disk count?

No. Enter only active members of the modelled array. A dedicated hot spare does not add usable capacity until it replaces a failed member. Filesystem and controller reserves are also excluded.

RAID Calculator

Calculate theoretical usable capacity for RAID 0, 1, 5, 6, 10, 50, 60 and independent JBOD disks using one disk size. Compare the same disks under every level, choose equal groups for RAID 50 or 60, and switch between decimal TB and binary TiB. Failure counts distinguish the number tolerated in any placement from the higher count possible across different mirrors or parity groups. Filesystem/controller overhead, hot spares, speed and rebuild risk are not calculated.

RAID 5 · 4 disks · 4 TB each12TBusable of 16 TB raw · any 1 disk can fail, whichever one it is.

Usable 12 TB (75%)Parity 4 TB (25%)

Usable = (4 − 1) × 4 TB = 12 TB

Level

Disks

Disk size

Show results in

One equal disk size; active array members only. Hot spares and filesystem/controller overhead are excluded. The model accepts up to 1,024 disks.

Usable10.91 TiB, the same bytes counted the other way
12 TB
Lost to protectionparity, 25% of the 16 TB raw total
4 TB
Disks that can failAny 1 disk can fail, whichever one it is.
1 of 4
Minimum disksfor RAID 5, and you have 4
3
Reads and writes
Reads can use striped data. A partial-stripe write may require reading old data and parity before updating both; cache and full-stripe writes change the work.

A tebibyte is exactly 1,099,511,627,776 bytes, so 12 TB of usable capacity is 10.91 TiB. Some software displays binary capacity, sometimes labelled TB. Display conventions vary; filesystem and controller overhead are not included here.

Equal-size model: standard striped or mirrored layouts here use the smallest disk. Enter that size for the conventional estimate. Mixed-disk pools, unequal mirror pairs and vendor-specific layouts are not modelled.

The same disks under every level

4 disks of 4 TB, compared under each level. Pick a level name to switch to it. RAID 50 and 60 use the grouping that keeps the most capacity.

Usable	Lost to protection	Disks that can fail	Minimum disks
16 TB	0 TB (0%)	0	2
4 TB	12 TB (75%)	3	2
12 TB	4 TB (25%)	1	3
8 TB	8 TB (50%)	2	4
8 TB	8 TB (50%)	1 (up to 2)	4
RAID 50 needs at least 6 disks and you have 4. It stripes across RAID 5 groups, and the smallest group is three disks. Add 2 more disks, or choose RAID 5.			6
RAID 60 needs at least 8 disks and you have 4. It stripes across RAID 6 groups, and the smallest group is four disks. Add 4 more disks, or choose RAID 6.			8
16 TB	0 TB (0%)	0	1

RAID is not a backup. Disk redundancy does not create a separate historical copy. Deletion, corruption and shared hardware failures may still affect the array. The failure counts above assume healthy remaining disks and a functioning implementation; no rebuild-risk estimate is supplied.

RAID levels compared

N counts active disks, excluding hot spares. Formulae assume equal-size conventional groups; the failure figures assume healthy remaining disks and a functioning array implementation. Independent JBOD disks are modelled, not a specific spanned-volume format.

Level	What it does	Minimum disks	Usable capacity	Disks that can fail	Reads and writes
RAID 0	Striping, no redundancy	2	N × the smallest disk	None	Reads and writes can use multiple disks without parity. Actual throughput depends on the disks, controller and workload; this calculator does not estimate speed.
RAID 1	Mirroring	2	One disk	All but one disk	Reads may use different mirror copies. Writes update every copy; actual read and write speed depends on the implementation and workload.
RAID 5	Striping with distributed parity	3	(N − 1) × the smallest disk	1 disk	Reads can use striped data. A partial-stripe write may require reading old data and parity before updating both; cache and full-stripe writes change the work.
RAID 6	Striping with double parity	4	(N − 2) × the smallest disk	2 disks	Reads can use striped data. Writes maintain two parity values; the I/O cost depends on write size, cache and implementation.
RAID 10	Mirrored pairs, striped	4	N ÷ 2 × the smallest disk	1 disk, and up to one in every pair	Reads and writes can use different mirrored pairs. There is no parity calculation; actual speed and rebuild time are not estimated.
RAID 50	Striped RAID 5 groups	6	(N − groups) × the smallest disk	1 disk in every group	Reads and writes stripe over RAID 5 groups. A failed disk is rebuilt within its group; rebuild duration and risk are not estimated.
RAID 60	Striped RAID 6 groups	8	(N − 2 × groups) × the smallest disk	2 disks in every group	Reads and writes stripe over RAID 6 groups. Two parity values are maintained in each group; no speed or rebuild-risk estimate is supplied.
JBOD	Just a bunch of disks	1	Every disk in full	None; independent disks assumed	For independent disks, a read or write uses the disk holding that file. Spanned-volume behavior depends on the filesystem or controller.

Common questions

What is the difference between RAID 5 and RAID 6?: RAID 5 reserves one disk’s worth of distributed parity and tolerates any one disk failure. RAID 6 reserves two and tolerates any two. Four equal 4 TB disks therefore give 12 TB in RAID 5 or 8 TB in RAID 6 before filesystem and controller overhead. These are theoretical failure counts assuming healthy remaining disks and a functioning implementation. Unreadable sectors, rebuild duration and the probability of data loss are not estimated.
How much usable space does RAID 10 give?: This conventional RAID 10 model mirrors disks in pairs and stripes across the pairs, so eight equal 4 TB disks give 16 TB usable from 32 TB raw. It requires an even disk count of at least four. Any single disk failure is tolerated; up to four are possible only if each is in a different pair. Two failures in the same pair lose that pair. No transfer-speed or rebuild-time estimate is supplied.
Why does a 12 TB array show up as 10.9 TB?: Decimal TB and binary TiB count the same bytes with different divisors. One TB is 1,000,000,000,000 bytes; one TiB is 1,099,511,627,776. Thus 12 TB is about 10.91 TiB. Some software uses binary units but displays a TB label, while other software uses decimal units. The switch changes this display conversion only. Actual formatted capacity also depends on filesystem and controller overhead.
What happens if the disks are different sizes?: The conventional striped/mirrored calculations here use one common size, usually the smallest disk: a 6 TB disk beside 4 TB disks contributes 4 TB to that model. Unequal mirror pairs and vendor-specific mixed-disk pools may differ. Independent JBOD capacity is the sum of the disks, but this tool accepts one size; add mixed sizes separately. A spanned JBOD volume may lose access to more than the failed disk’s data.
What are RAID 50 and RAID 60?: RAID 50 stripes across equal RAID 5 groups and reserves one disk per group. RAID 60 uses equal RAID 6 groups and reserves two per group. Twelve 4 TB disks in three groups give 36 TB in RAID 50 or 24 TB in RAID 60. RAID 50 tolerates any one failure, or up to three if each is in a different group. RAID 60 tolerates any two, or up to six if no group loses more than two. This tool enumerates valid equal group counts.
Is RAID a backup?: No. Redundancy can keep an array available through specified disk failures, but it does not provide a separate historical copy of data. Deletion, corruption, ransomware and shared hardware failures can affect the array. This calculator compares capacity and theoretical disk-failure tolerance; it does not plan or verify backups.
Does this calculate RAID speed, IOPS or rebuild time?: No. The comparison explains how data, mirrors and parity are arranged, but supplies no throughput, IOPS, rebuild duration or unreadable-sector probability. Those require details about disks, controller, workload and software.
Does this reproduce Synology SHR, ZFS RAIDZ or a NAS vendor’s capacity calculator?: No. This is a conventional equal-size RAID model. SHR, RAIDZ, Btrfs, vSAN and proprietary controller or NAS layouts may have different allocation, metadata, reserve and failure behavior. Disk size is entered in decimal GB or TB; the result can be displayed in TB or TiB.
Are hot spares included in the disk count?: No. Enter only active members of the modelled array. A dedicated hot spare does not add usable capacity until it replaces a failed member. Filesystem and controller reserves are also excluded.

Conventional equal-size capacity formulae, rounded for display. Metadata, filesystems, hot spares, vendor layouts, performance and rebuild risk are not modelled.