Hardware failures, power failures, and human errors can corrupt information that your computer needs to start Windows NT.
Recovery is easier if:
On computers running Windows NT Server, you can purchase disk subsystems that maintain redundant information on the disks. The redundant information is either parity information, or a complete, separate copy of the data. Having redundant information makes the disk subsystems fault tolerant, meaning that you can continue to access the data when one disk in a fault-tolerant configuration fails. This fault-tolerant technology is called RAID, for Redundant Array of Inexpensive Disks. A disk subsystem that implements fault tolerance using RAID technology is also called a RAID array.
Windows NT Server implements two of the fault-tolerant configurations in software: mirror sets, and stripe sets with parity. For a description of mirror sets and stripe sets with parity, see Chapter 4, "Planning a Reliable Configuration."
For complete information about RAID, RAID terminology, and RAID arrays, see The RAIDbook, A Source Book for Disk Array Technology. The RAID Advisory Board in St. Peter, MN, publishes this book. The June 1995 edition is the basis for RAID information in this chapter. The RAIDbook, A Source Book for Disk Array Technology also contains information about parity.