Different types of RAID

RAID (Redundant Array of Independent Disks) is a technology that helps to store data in a safer way or help you to access to it more efficiently.
This is achieved by using several hard drives.

The main raid arrays

Raid 0:
The main purpose is to increase performances for reading and writing data. Your files are splitted among several hard drives (2 minimum). The big issue of raid 0 is that it has no tolerance for any of the drive failure. If one hdd crashed, then the loss of data is immediate. More info for RAID 0 data recovery on this link.

Raid 1:
Also called mirroring. Datas are written on both drives for example but could be 3 or 4 drives.
Performances can be good and on a safety point of view it is strong. If one hdd crashed, there is no loss of data.
Still we see in our lab some Raid 1 with 2 drives. Most of times because of hdd felle on the ground or manages to be dead at the same time.

Raid 5:
You need a minimum of 3 disks to set up a raid 5. The array can handle the crash of a single hard drive maximum without loss of data. Basically your files are splitted among disks with distributed parity than allow reconstruction of the failed drive by calculating the loss data out of this parity element.

The biggest advantage of RAID 5 is a equal reading performance as raid 0 with a much better security level since the crash a a single drive does not result in any data loss.

Speaking about inconvenient, a raid is often composed of big hard drives like 1To or 2To that will considerably affect the time you will need to rebuild the array..could be several days.

More info about RAID 5 data recovery on this link.

The other types of RAID

Raid 2:

Not used anymore or little used since it has low performances

Raid 3 and 4:

Those 2 are quite similar though raid 4 has better performances.
In a few words the store data on sevaral disks and parity is stored on a dedicated disk so if one hard drive crashed data can be rebuilt.

Raid 6:

It is a close idea to raid 5 except that it can resist the failure of several hard drives. Basically this raid uses more than a single parity. If a raid uses a single distributed parity when one hdd crashes, then data are very vulnerable (same as raid 0) until the dead hdd has been changed and the array rebuilt. In the case of raid 6, there is no risks of data loss while recovering from a single hard drive crash.
The drawback is that rebuilding raid 6 takes much longer time.


Increases abilities of raid 6.

RAID DP or Dual parity:

Very close to raid 6 but with fixed hard drives for parity. You find this set up in NAS (Network Attached Storage).

Raid 01 or 0+1

A minimum of 4 disks is required. It increases performances but security is average since the failure of one hard drive will kill one array.
Basically this is a mirror of a raid 0. Complexity increases with such set up. You can find raid 1+0 which would be a stripped of mirrors. This set up is better because it can resist the loss of several hard drives as long as one full mirror is not dead.

RAID 51 or 5+1 or 53

Requires a minimum of 6 disks and 2 arrays. It is a mirror of raid 5. Expensive and little capacity for a big number of drives.

RAID 5E or Enhanced

Raid 5 like and improvements


Kind of RAID 5E with better rebuilding time.

RAID 5DP or Dual Parity

Similar to raid 6. Oftenly used in NAS (Network Attached Storage).

RAID TP (Triple Parity RAID)

Similar to raid 6 with 3 parity distributed. If 3 drives crashes it still works.

RAID 1.5

Evolution of raid 1. Better performances since data are read from both drives.


Improvement of raid 3 and property of Storage Computer Corp. Expensive and performant.


Matrix RAID (from Symmetrix)

Ability to combine raid 0 and 1 on 2 hdd for example.


Mirroring system

Sun RAID Z and Z2

Improvement of RAID 5


A netgear product using mainly raid 1 and 5 and allowing you to play between both..

In terms of data recovery, if we are talking raid with a physical failure, we are speaking about an longer time than a single drive.
As you have read before, one reason is due to the raid recontruction time and complexity.

Another one is due to the large capacity drives used nowadays. For drive over 1To, recovery is more complex and long.