# What is my RAID capacity? Computes the usable capacity, storage efficiency, and number of disk failures survived for a RAID 0, 1, 5, 6, or 10 array of identical disks. - Page: https://www.acalculator.org/logic/raid-calculator - JSON spec: https://www.acalculator.org/logic/raid-calculator.json - Version: b6e84fcfae15 ## Default answer Example with the default inputs (RAID level RAID 5, Number of disks 4, Size of each disk 4 TB): 4 disks of 4 TB in RAID 5 give 12 TB of usable space. Disk failures survived: 1. ## Inputs | Key | Label | Description | | --- | --- | --- | | level | RAID level | How the array spreads and protects data: RAID 0, 1, 5, 6, or 10. | | disks | Number of disks | How many disks are in the array, all the same size. | | size | Size of each disk | The capacity of one disk, as printed on the label (1 TB = 1,000,000,000,000 bytes). | ## Outputs | Key | Label | Description | | --- | --- | --- | | usable | Usable capacity | The space left for your data after parity or mirror copies, in decimal terabytes. | | usableTiB | Usable capacity in TiB | The same usable space in tebibytes (2⁴⁰ bytes), the unit most operating systems show as “TB”. | | raw | Total raw capacity | The size of all the disks added up. | | protection | Used for protection | Raw capacity that holds parity or mirror copies instead of data. | | efficiency | Storage efficiency | Usable capacity as a percent of the raw capacity. | | tolerance | Disk failures survived | How many disks can fail at the same time, in any combination, without losing data. | | toleranceBest | Most failures survived (RAID 10) | How many disks can fail without data loss when each failed disk is in a different mirror pair. | ## Method Usable = RAID 0: N × S; RAID 1: S; RAID 5: (N − 1) × S; RAID 6: (N − 2) × S; RAID 10: N ÷ 2 × S, for N disks of size S. ## Assumptions - All disks are the same size. In a mixed array, each disk counts as the smallest one. - RAID 10 uses two-way mirror pairs and needs an even number of disks. RAID 1 mirrors the same data on every disk. - Capacities are before the file system, hot spares, and the controller’s own metadata take their share. - Disk sizes use decimal units (1 TB = 10¹² bytes), as drive makers label them. ## Worked examples 1. level = 5, disks = 4, size = 4,000,000,000,000 gives usable = 12,000,000,000,000, usableTiB = 12,000,000,000,000, raw = 16,000,000,000,000, protection = 4,000,000,000,000, efficiency = 75%, tolerance = 1. Source: hand calculation in content.mdx: (4 − 1) × 4 TB = 12 TB = 10.91 TiB (raw values in bytes); SNIA dictionary, RAID 5 (n data + 1 parity). 2. level = 6, disks = 8, size = 10,000,000,000,000 gives usable = 60,000,000,000,000, efficiency = 75%, tolerance = 2. Source: hand calculation in content.mdx: (8 − 2) × 10 TB = 60 TB; SNIA dictionary, RAID 6 (survives any two failures). 3. level = 10, disks = 6, size = 2,000,000,000,000 gives usable = 6,000,000,000,000, efficiency = 50%, tolerance = 1, toleranceBest = 3. Source: hand calculation in content.mdx: 6 ÷ 2 × 2 TB = 6 TB; SNIA dictionary, RAID 10 (RAID 0 over RAID 1 pairs, one failure per pair). 4. level = 1, disks = 3, size = 2,000,000,000,000 gives usable = 2,000,000,000,000, efficiency = 33.333333%, tolerance = 2. Source: hand calculation in content.mdx: every disk holds the same data, so 2 TB usable and 3 − 1 = 2 failures. 5. level = 0, disks = 3, size = 500,000,000,000 gives usable = 1,500,000,000,000, efficiency = 100%, tolerance = 0. Source: hand calculation in content.mdx: 3 × 500 GB = 1.5 TB; SNIA dictionary, RAID 0 (no redundancy). ## FAQ ### How do I calculate RAID 5 capacity? Multiply the size of one disk by the number of disks minus 1. RAID 5 spreads one disk’s worth of parity across the array. Four 4 TB disks give (4 − 1) × 4 TB = 12 TB of usable space, and the array survives one disk failure. ### What is the difference between RAID 5 and RAID 6? RAID 5 keeps one disk’s worth of parity and survives one failed disk. RAID 6 keeps two and survives any two failed disks at the same time. RAID 6 costs one more disk of capacity, which matters most with large disks that take a long time to rebuild. ### How many disks can fail in RAID 10? At least one. RAID 10 stripes data over mirror pairs, and each pair survives losing one of its two disks. If every failed disk is in a different pair, up to half the disks can fail, but two failures in the same pair lose data. ### Why does my operating system show less space than the calculator? Drive makers count 1 TB as 1,000,000,000,000 bytes, but Windows and many other systems divide by 2⁴⁰ and still label the result TB. 12 TB is 10.91 TiB, which Windows shows as 10.9 TB. The file system and the RAID controller’s own metadata take a little more. ### Is RAID a backup? No. RAID keeps the data available when a disk fails, but it copies deletions, corruption, and ransomware to every disk at once, and a fire or theft takes the whole array. Keep separate backups as well. ### What happens if my disks are different sizes? Standard RAID 1, 5, 6, and 10 use the same amount of space on every disk, so each disk counts as the smallest one and the extra space on larger disks is unused. Enter the size of the smallest disk to see the usable capacity. ### Which RAID level should I choose? RAID 0 gives all the space and speed but no protection. RAID 1 mirrors two disks for simple protection. RAID 5 gives the most space with protection against one failure, RAID 6 protects against two, and RAID 10 protects well and rebuilds fast at the cost of half the space. ## Sources - SNIA Online Dictionary, RAID 0. https://www.snia.org/education/online-dictionary/term/raid-0 - SNIA Online Dictionary, RAID 1. https://www.snia.org/education/online-dictionary/term/raid-1 - SNIA Online Dictionary, RAID 5. https://www.snia.org/education/online-dictionary/term/raid-5 - SNIA Online Dictionary, RAID 6. https://www.snia.org/education/online-dictionary/term/raid-6 - SNIA Online Dictionary, RAID 10. https://www.snia.org/education/online-dictionary/term/raid-10 - NIST SP 811 and IEC 80000-13: 1 TB = 10¹² bytes, 1 TiB = 2⁴⁰ bytes.