Use this calculator to easily calculate RAID capacity, disk space utilization, cost per usable TB, read/write efficiency (I/O operations per second improvement) and more. Supported levels are: RAID 0, RAID 1, RAID 1E, RAID 4, RAID 5, RAID 5E/EE, RAID 6, RAID 10, RAID 50, and RAID 60.
What is RAID?
RAID (stands for Redundant Array of Independent Disks, originally Redundant Array of Inexpensive Disks) is a data storage virtualization technology that combines multiple physical disk drive components into one or more logical units for the purposes of data redundancy, performance improvement, or in certain cases: both. The term was coined in a paper by Patterson at.al. in 1987-1988 when they argued that an array of cheap magnetic disks can outperform the top performing mainframe drives of the time.
In a RAID data is distributed among the drives (could be HDD or SSD) in one of several ways called RAID levels. The particular configuration depends on the required level of performance and redundancy. Each RAID level provides a particular balance across several variables: reliability, availability, performance, and capacity. The techniques to achieve that are: mirroring, in which identical data is copied onto more than one drive; striping, which partitions each drive's storage space into units ranging from a sector up to several mb; parity - in which information is striped across each drive, allowing the RAID to continue working even if one drive were to fail. Parity uses the XOR operation to allow information to be restored in case of a drive failure. Check the graphs below the levels table for visuals.
This RAID calculator outputs metrics such as redundancy factor, price per usable TB/GB, as well as read and write speed gains, which can help you decide if the selected configuration is right for your particular case. RAID levels greater than RAID 0 provide protection against unrecoverable sector read errors, as well as against failures of whole physical drives. A detailed list of RAID levels follows.
List of RAID Levels (configurations)
|RAID level||Description||Additional information|
|RAID 0||Striping only||This is the simplest RAID configuration possible, without mirroring or parity so there is no redundancy or recovery capacities. The performance of both reads and writes approaches the sum of throughputs of every drive in the set and is a big benefit of this spanned configuration.|
|RAID 1||Mirroring only||All data is written identically to two drives (mirrored set), so any read request can be serviced by any of the drives, achieving redundancy. A hypothetical read speed of such a configuration is the sum of throughputs of the individual drives. Write speed is negatively affected as the slowest drive limits the performance of the array.|
|RAID 1E||Striping with mirroring||Each written stripe is mirrored to one of the remaining disks in the array. Has better random read performance than RAID 1, even in a degraded array. (E stands for Enhanced) RAID 10 is usually preferred over RAID 1E these days.|
|RAID 2||Bit-level striping with dedicated parity||This RAID type is rarely used and thus not supported by our calculator.|
|RAID 3||Byte-level striping with dedicated parity||This RAID type is rarely used and thus not supported by our calculator.|
|RAID 4||Block-level striping with parity||The advantage of RAID 4 over 2 and 3 is I/O parallelism. In the latter, a single I/O read requires reading the whole group of data drvies, while in RAID 4 one I/O read does not have to spread across all drives, which improves performance of small transfers.|
|RAID 5||Block-level striping with distributed parity||Unlike RAID 4, parity information is distributed among the drives, requiring all but one to operate. RAID 5 might be susceptible to system failures due to long rebuild time during which a second drive might fail, which is why some advise against its usage.|
|RAID 5E / 5EE||RAID 5 with an integrated hot spare||The hot spare is an active part of the block rotation scheme, spreading I/O across all drives and reducing the load on each drive, which results in increased performance. The spare cannot be shared among multiple arrays. E stands for Enhanced.|
|RAID 6||Block-level striping with double distributed parity||Double parity provides fault tolerance for up to two failed drives, making larger arrays practical, especially for high-availability systems with big drives that might take a long time to rebuild. As with RAID 5 the entire array will perform poorly until a failed drive is restored. Larger drive capacities and larger array sizes is where RAID 6 really shines.|
|RAID 10||Mirroring without parity, and block-level striping||This configuration consists of a striped set from a series of mirrored drives: RAID 1 + RAID 0, essentially. As long as no one mirror loses all its drives, the array will survive. Good when both high performance and high security is required. Should not be mistaken with RAID 01.|
|RAID 50||Block-level striping with distributed parity, and block-level striping||This configuration combines the straight block-level striping of RAID 0 with the distributed parity of RAID 5. In essence, it is a combination of multiple RAID 5 groups with RAID 0. One drive from each RAID 5 array may fail without data loss, so a RAID 50 array with three RAID 5 sets can tolerate a total of 3 drive failures. The improvement over RAID 5 is in better performance, especially for writes, and higher fault tolerance. Hot spares can be used as well.|
|RAID 60||Block-level striping with double distributed parity, and block-level striping||This is a RAID 0 array striped across RAID 6 elements, making it require at least 8 disks (2 sets of 4 disks) at minimum. It offers great reliabilty, but at a heavy cost in terms of usable capacity as percent of overall disk capacity. You can lose between 12% and 50% of your disk space to parity information - the more disks you add to each individual RAID 6 set, the higher the percentage of usable capacity. Writes take a performance hit, but reads are boosted.|
Striping, mirroring and parity - graphs:
How to calculate RAID capacity?
In order to calculate the capacity of a disk array, you need to know the peculiarities of each configuration, as some parts of it will limit the usable capacity of the array, like mirroring and parity. Then, you need to know the number and size of the remaining disks and multiply them together. It is that simple, and it's even easier using our calculator. No matter if you need a RAID 10 calculator, RAID 5 calculator, RAID 6 calculator, etc. - it's all in one.
Disk space versus usable disk space (capacity)
In a RAID, mirroring and parity decrease the usable disk space. This is the cost to have advantages like fault tolerance and high availability. The usable disk space can be as low as 50% of the total disk space you buy, so beware about the trade-offs involved in using RAID and study each configuration carefully. The RAID calculator will greatly assist you in this task.
Drive Size: TB vs TiB, GB vs GiB
In our RAID calculator you can select between four types of storage units: classic binary terabytes and gigabytes and SI terabytes and gigabytes. In order to differentiate between the two, the International System of Units (SI) made the horrible decision to start using existing terminology to refer to something else. Thus the terabyte (TB) became 1000 GB, instead of 1024. So, they needed a name for the 1024GB terabyte, so they chose "tibibyte" with a symbol "TiB". The story is the same with gigabyte and gibibytes (GiB).
What you should know is that most operating systems report file sizes and disk sizes in binary TB and GB (1TB = 1024GB, 1GB = 1024MB) while disk manufacturers denote the size of their drives in SI TB and GB (1TB = 1000GB, 1GB = 1000MB). Make sure to account for that insanity while planning your RAID. In our RAID calculator we support all four.
Do I still need a backup?
Absolutely! RAID is no replacement for backup. Even though some RAID levels provide data redundancy, that doesn't mean it should be used as backup of your critical files. While a RAID protects you against drive failure, it does not protect you against errors (human or otherwise), file corruption, malicious actors or RAID controller failure.
 Patterson D., Gibson G. A., Katz R. (1988) "A Case for Redundant Arrays of Inexpensive Disks (RAID)" Proceedings of the 1988 ACM SIGMOD international conference on Management of data 17:109-116
 Arpaci-Dusseau R. H., Arpaci-Dusseau A. C. (2015) "Operating Systems: Three Easy Pieces", chapter "Redundant Arrays of Inexpensive Disks (RAIDs)"
Cite this calculator & page
If you'd like to cite this online calculator resource and information as provided on the page, you can use the following citation:
Georgiev G.Z., "RAID Calculator", [online] Available at: https://www.gigacalculator.com/calculators/raid-calculator.php URL [Accessed Date: 17 Oct, 2018].