RAID, whilst relatively simple (in most applications at least), is one of the more important storage related developments of the recent past.
TRC Data Recovery are experts in RAID Recovery.
RAID is an acronym that has two different interpretations. Originally the acronym was Redundant Array of Inexpensive Disks, coined by the developers led by David A. Patterson of the University of California, Berkeley. It soon became clear that (in the early days at least), RAID was in no way an inexpensive technology to implement and thus a decision was made to amend the acronym slightly to Redundant Array of Independent Disks which seemed a more appropriate term.
In the early days of RAID things were relatively simple, there were stripes, mirrors and stripe sets that had some redundancy. Through the years a number of different forms of RAID array have hit the market, some of these have experienced some commercial popularity and success, others have all but disappeared from the market due to better alternatives existing. We will now look at some of the more common array types:
RAID 0 is a method of striping data across two or more storage devices. The data to be written is split up into small chunks, either by a hardware RAID controller or through a software implementation. This data is then written to each of the member devices. Doing so speeds up the storage but actually increases the risk of data loss exponentially as the number of component disks increases.
Because of this RAID 0, despite containing the term RAID in its name is not an official RAID type as there is no redundancy (does AID have the same ring to it?). This is a large issue when a single device fails as all of the data is lost.
Read about: RAID 0 Data Recovery
RAID 1 consists of a minimum of two storage devices that each have the same set of data written to them simultaneoulsy. In such cases the data is essentially backed up to all drives in the array. RAID 1 is often referred to as mirroring due to identical, mirror image that is created on each member disk. Unlike most other array types RAID 1 carries no real speed advantage over a single disk configuration. As such it is often used in lower end systems in a two disk guise purely to provide some redundancy against one of the member hard drives failing.
Note: RAID 0 should not be confused with a 'real' backup solution as any corruption to the data or the operating system will be duplicated across all of the member devices within the array. If something goes wrong and no external backup or snapshot was taken, the risk of major data loss still exists.
Read About: RAID 1 Data Recovery
RAID 5 is perhaps the most common form of RAID array in use today. RAID 5 offers a speed bump whilst retaining redundancy against a single member drive in the array failing.
Like RAID 0, RAID 5 arrays stripe the data across the component drives which increases the performance of the whole storage system. The redundancy aspect of RAID 5 comes in to play thanks to the use of distributed parity. An exclusive disjunction or XOR system uses a simple truth table to determine a calculation of the XOR of the striped blocks. This parity block is then distributed amongst the member disks, swapping where it is written to on each stripe cycle (although some hardware RAID controllers write the parity in larger blocks before moving to another member disk).
RAID 5 often offers the best compromise between redundancy and speed, whilst using the lowest number of physical disks.
Read About: RAID 5 Data Recovery