An acronym for Redundant Array of Independent Disks. Sometimes you will see the word Inexpensive in place of Independent, but the basic point is the same: Multiple drives doing the work of one drive, but with redundancy. The term was coined in 1987 by researchers at the University of California at Berkeley to describe a series of redundant architectures used in fault-tolerant disk arrays (RAID levels 1 through 5). It was originally developed as a way to organize small format disk storage devices to dramatically increase input/output (I/O) speed and improve data availability. At its simplest, a RAID-1 array consists of two drives that store identical information. If one drive goes down, the other continues to work, resulting in no downtime for users. RAID-1 isn’t a very efficient way to store data, however. To save disk space, RAID-3, -4, and -5 “stripe” data and parity information across multiple drives (RAID-3 and -4 store all parity data on a single drive). If a single disk fails, the parity information can be used to rebuild the lost data. Unfortunately, there is a performance trade-off: depending on the RAID type used, a RAID will be slower than a single drive at either reading or writing data. In demanding audio or video applications, however, a properly configured RAID system can actually increase bandwidth (speed) while improving reliability.