The following methods provide
fault tolerance for hard-disk systems:
Redundant Array of Independent Disks (RAID)
a drive means designating a hard-disk drive in the computer as a duplicate to
another specified drive. The two drives are attached to a single disk
controller. This fault tolerance feature is provided by most of the network
operating systems. When any data is written into the drive, the same data is
also written to the drive designed as the mirror. If the drive fails, the mirror
drive is already online, and because it has supplicate information, the users
won’t realize that a disk drive in the server has failed. The NOS notifies the
admin that the failure has occurred. On the other side if the disk controller
fails neither drive is available.
Duplexing also saves data to a mirror drive; the only major difference
between duplexing and mirroring is that duplexing uses two separate controllers.
Hence duplexing not only provides redundant disk but also redundant controller.
Duplexing provides fault tolerance even if a controller fails.
Disk striping breaks up the data that are to be
saved to the disk into small portions and sequentially writes the portions to
all disks simultaneously in small areas called strips. These strips maximize
performance because all of the read/write heads are working constantly.
(Redundant Array of Inexpensive Disks)
uses an array of less-expensive hard disks and provides several methods for
writing tot hose disks to ensure redundancy. RAID has seven levels; each level
is designed for specific purposes.
RAID 0 is the commonly used disk. This method is the fastest because all
read/write heads are constantly being used without the burden of parity or
duplicate data being written. This RAID level improves the performance; it does
not provide fault tolerance.
also commonly used disk. This level uses hard disks, one mirrored to the other.
RAID 1 is the most basic level of disk fault tolerance. If the first hard disk
fails, the second hard disk automatically takes over. The parity or
error-checking information is not stored. Rather the drives have duplicate
information. If both the drives fail a new drive must be installed and
configured. This level provides fault tolerance.
level individual bits are stripped across multiple disks. Multiple redundancy
drives in this configuration are dedicated to storing error-correcting code.
level data is striped across multiple hard drives using a parity drive. The data
are striped in bytes and not in bits as of RAID 2. This configuration is popular
because more data is written and read in one operation that increases overall
is similar to RAID 2 and 3 expect the data is striped in blocks, which
facilitates fast reads from one drive. This is not popular implementation.
This level is commonly used; at this level the data and parity are
striped across three or more drives. This allows fast reads and writes. This
works well if one disk fails.
RAID 6 is similar to RAID 5. It is less popular. This level uses RAID 5
as a basis but duplicates the parity information, saving the second copy on a
different drive from one on which the first copy was saved.