RAID - An Overview
Submit Articles Back to Articles
RAID stands for "redundant array of independent disks," or sometimes
"redundant array of inexpensive disks." RAID refers to a hardware arrangement
that allows for the storage of data in multiple places on two or more disk
drives used in combination. The use of RAID can enhance the performance of a
system by allowing input and output procedures to overlap, and has the
additional advantage of providing fault tolerance. Fault tolerance refers to the
level of a system's capability to respond to problems, such as hardware or
software failure. RAID improves fault tolerance because it extends the range of
time that a system can be expected to function reliably between failures, which
is commonly referred to as the mean time between failures (MTBF).
There are at least nine established levels of RAID, each using different
combinations of techniques and having their own benefits and disadvantages. To
the operating system, which manages the application programs of a computer, RAID
is perceived as a single hard disk. A defining element of RAID is the use of
disk striping, which solves the problem of slow data transfers that limit
performance in systems employing a single drive. Striping divides larger files
into small pieces that are stored across multiple drives such that the files can
be retrieved efficiently as the drives work in parallel.
There are two data redundancy methods used in RAID, mirroring and parity.
Disk mirroring is a technique that is used in some RAID levels, particularly
RAID Level 1 but also in other levels which incorporate Level 1. In disk
mirroring, at least two drives are used to simultaneously store the same data,
adding some level of data protection to the array. In the case that one of the
mirrored drives fails, the system can automatically switch to using the
functional disk without downtime or data loss. Though disk mirroring does not
replace periodic back ups for critical data altogether, it is being increasingly
implemented in sectors such as the financial industry as an alternative to the
highly inefficient and basically antiquated tape based back ups. Parity refers
to a technique which uses given data to compute an extra piece of data, and
stores the multiple pieces of data across multiple drives. The premise is that
if one piece of data is lost, it can be recreated using the remaining data.
Parity provides a cost advantage over mirroring, as many disks are not required.
About the Author
Stephen J. Richards has 25 years experience in Data Management and
Information Technology. This information is provided as a public service by Neon
Enterprise Software, a leading provider of
For more information, please visit
Follow us @Scopulus_News
Article Published/Sorted/Amended on Scopulus 2008-05-16 23:58:02 in Computer Articles