RAID Calculator – Calculate Usable Capacity and Fault Tolerance

Buy six 4 TB drives and you do not get 24 TB of usable storage. What you get depends on the RAID level you choose. Enter the number and size of your drives, pick a RAID level and see the usable capacity and how many drive failures the array can survive, all before you place an order.

RAID levels at a glance

In the table, n is the number of drives and s is the size of the smallest drive.

RAID level

Minimum drives

Usable capacity

Drive failures survived

Typically used for

RAID 0

2

(n × s)

None

Temporary or scratch data where speed matters and loss is acceptable

RAID 1

2

s (one drive's worth)

1 in a two-drive mirror

Small servers, operating system drives

RAID 5

3

(n − 1) × s

1

General file storage that is mostly read

RAID 6

4

(n − 2) × s

2

Large drives, archives, backup targets

RAID 10

4

(n ÷ 2) × s

At least 1; more if the failed drives are in different mirrors

Databases, virtual machines, write-heavy workloads

How to calculate RAID capacity

  • Enter the number of drives in the array.
  • Enter the capacity of each drive. If your drives differ in size, use the smallest, because the array treats every drive as that size.
  • Select the RAID level.
  • Read the usable capacity and fault tolerance.
  • Compare other levels before you decide.

Why capacity is not the only thing to compare

  • Rebuild Risk: When a drive fails, the array rebuilds onto a replacement by reading all the remaining drives. On large modern drives that can take many hours or even days, and the array has little or no protection while it runs. If a second drive fails, or an unreadable sector turns up in RAID 5 during that window, data can be lost. This is why RAID 6 or RAID 10 is commonly preferred for large drives and important data.
  • Write Performance: Parity RAID needs extra work for every small write. Roughly, RAID 10 needs 2 disk operations per write, RAID 5 needs 4 and RAID 6 needs 6. For databases and virtual machines, that difference is noticeable.
  • Hot Spares: A spare drive that is already installed lets the rebuild begin immediately instead of waiting for someone to arrive with a replacement.

Planning tips

  • Use drives designed for NAS or server workloads in arrays that run all day.
  • Keep at least one spare drive of the same size or larger.
  • Set up alerts so that a failed drive is noticed straight away.
  • Document the configuration, including level, drive models and controller settings.
  • Buy drives in staggered batches where practical, so they do not all wear out at the same time.

Frequently asked questions

RAID 0 gives the most, but it offers no protection at all. Among the levels that do protect your data, RAID 5 is the most space-efficient, especially as you add drives.

RAID 6 can survive any two. RAID 10 can survive two only if they are in different mirrored pairs.

For smaller drives and less critical data, RAID 5 can be reasonable. With large drives or business-critical data, RAID 6 is generally the safer choice because of its second parity drive.

 Yes, for small servers and boot drives where simple mirroring is enough. It uses half of your raw capacity.

Usually, but the array uses the smallest drive's size for each member, so the extra space on larger drives is wasted.

No. RAID improves availability. A backup protects against data loss.

Need help choosing, building or maintaining server storage? See our infrastructure and Windows Server services.