Step to take to make data recovery from RAID disk arrays

That will depend on the type of RAID system you are using.

So what do the different RAID types represent?

Let us first define what RAID stand for and what it is.

The acronym of RAID stands for "Redundant Array of Inexpensive Disks" which was coined in 1987. It also sometimes stands for "Redundant Array of Independent Disks"

Basically RAID's are arrays of disks which working as a single hard disk viewed from the pc computer or the server.



The main advantages for using RAID are:

• Gives high reliability in case one disk fails.
• Provide protection and high uptime for data with uninterrupted data access.
• Increases performance by letting disks work in parallel.

There are various RAID types with different relevance, if you consider making recovery from RAID arrays. Not all advantages are applicable for all types of RAID's.

RAID 0 uses stripping. Stripping means that when data are written the data are split-up in parallel during the writing transfer over the disks. RAID 0 gives you high throughput and is good for things with high disk utilizations. One example is if you edit videos, images or audio. The downside is that if one disk fails it is gone and if you want to make a disaster data recovery from the RAID 0, then you have to send the RAID disk array to a data recovery company for service and restoration.
You need to have an external backup strategy in place if you work with RAID 0.

RAID 1 uses disk mirroring of two hard disks or doublets that duplicate data storage.
If one disk fails you can still run the system while not recovering data from RAID arrays, but you have to replace the faulty hard disk with a new one as quick as possible.
One advantages of a system using RAID 1 mirroring is the high level of redundancy that is achieved. If you store lets say 50GB you need at least two 50GB which both store the same data.
This system is best for computers or severs that need high level of availability and protection. This can be the case if you run applications in a business working environment with software applications such as accounting.

RAID 10 uses stripping among two pairs. RAID 10 uses combinations of the techniques of RAID 0 and RAID 1. The data are stripped among two pairs and each pair uses disk mirroring. In this case redundancy and high throughput is achieved. To have a storage capacity of 50GB at least four 50GB disks must be used.
If one set of mirror disk crash then the other will still work. You don't need to make recovery from the data RAID disks. You just have to make a replacement of at least one of the disks.
RAID 10 is used for system that need high availability and fault tolerance. This can be in high demand database applications.

RAID 0+1 uses mirroring across twin pairs of drives as opposed to RAID 10 which uses stripping across twin pairs. Each pair uses stripping across the pairs drives in the configuration. If one disk fail and one of the pair goes down the RAID system becomes a RAID 0 without fault tolerance.
This configuration is useful of you need high throughput but can accept long downtimes as is often the case of you work with video intensive applications.

RAID 5 uses stripping and parity checks. This is done with a minimum of three drives using a rotating parity array. The stripping and parity check is mixed over the three drives. The result is that if one disk fails the system will still work. This is the case without making disk recovery from RAID. Before the failed disk is replaced the computer can still be running.
This configuration is versatile and is good for many software applications such as databases, email servers and for Intranet.