A raid array is several disks that are independent of each other, which are interconnected via channels operating at a fairly high speed. The system perceives this union as one single disk array. The distribution of read and write processes is controlled by a controller.
Raid array is used in order to maximize the write speed or reliability of disk storage. The solution of these issues can significantly increase the overall efficiency of the entire system.
Initially, the creation of a RAID array involved the integration of low-cost drives. This initiative was clearly reflected in the name itself (the abbreviation meant the following: "an excess array of inexpensive drives"). But subsequently, this technology, which has established itself as a quality and reliable, has changed. A modern raid array is created, as a rule, on the basis of rather expensive equipment. This, in turn, was again reflected in the name (today the abbreviation means "an excess array of independent disks").
The development of the information society has greatly influenced the requirements that are put forward for the hardware, including information storages.
Therefore, since the raid array was created, several different sub-technologies have appeared that were developed to solve specific narrowly targeted problems.
All the variety of tasks solved by computers requires equally diverse technological solutions. Types of raid arrays:
- RAID 0. In the simplest implementation, two drives are required. The incoming information is written in blocks in turn on both discs. Reading involves the reverse process. This significantly speeds up recording, but does not provide sufficient reliability, since an error on one of the disks leads to data loss in the entire array.
- RAID-1. In the simplest implementation, two drives are required. Information received on the discs is recorded on them in full. This slows down the recording process, but, thanks to one hundred percent redundancy, provides a high level of reliability. The main advantage of this development is its high fault tolerance and fast reading (since it is possible to read from two disks at the same time).
- RAID-5. Fault tolerant array. The simplest implementation requires three drives. Any recording of information occurs by breaking the data stream into blocks, each of which is assigned a checksum. In case of data loss on one of the disks, they can be restored using checksums recorded on the other. Recovery requires the allocation of large capacities, so most often this implementation is used on server hardware.
The fifth array raid is the most common. This is due to its reliability and high speed of information processing, since one array can simultaneously carry out several independent processes of writing and reading information.