RAID controller explained...

A RAID controller quite simply controls all the data on your array of hard drives, so that they work as a logical unit. By arranging the data on multiple physical hard drives otherwise known as redundancy, it has the ability to improve the RAID’s performance and protect the information in the event of data loss. RAID controllers are either hardware or software biased and some can integrated into the motherboard. Hardware based controllers are normally PCI cards which are designed to take specific hard drive interfaces. Software controllers are hosted on the CPU and provide similar functionality, but their performance is often limited.

RAID controllers that came as a card were often not reliable. Thankfully these days, RAID controller failure is quite rare. However we do see more embedded RAID controller failures as opposed to discrete controller failures. If your RAID controller fails, replacing it is not normally an option as it will try to re-initialise the disks. Attempts to rebuild the array will usually use default settings. When we receive a RAID for data recovery, the user has normally tried all these options in a desperate attempt to recover their data. Unfortunately this has normally destroyed the RAID parameters making data recovery much more difficult.

What happens if your RAID controller fails? Generally the data cannot be accessed in its original configuration and user files become inaccessible. To understand why, we must examine how data is stored across multiple disks. There are many different RAID levels of configuration to increase either performance of security. The data is then distributed across the hard drives in different ways depending on the level required to balance the user’s needs. For instance RAID 0 or ‘striping’ distributes data equally amongst all disks, meaning that if one disk fails the entire volume is lost with reduced possibilities for data recovery. At the other end of the scale, RAID 5 offers block level striping with distributed parity. This means that one disk can fail and the inherent ‘redundancy’ will automatically recover all the data within the controller. However if two or more disks fail, then specialist data recovery services will be necessary.

The RAID controller distributes all this data. If it fails the magnetic data remains unaltered, but that information simply cannot be interpreted. It then becomes necessary to re-interpret that data by calculating the data parameters and reverse-engineering the user’s information. More often than not this is a time consuming exercise that relies on a significant amount of cryptanalysis.