Fault Tolerance - Network+ Certification Tutorial

Mirroring: RAID 1 - Duplicates a partition on another physical disk with one data channel, 2 drives, 1 used for data and 1 for parity. Its advantages include fault tolerance. Disadvantages include it's expensive and requires twice as much disk space.

Duplexing: RAID 1 - Duplicates a partition on another physical disk that is connected to another Hard Drive Controller using two data channels simultaneously, two data cables and two DASD, 1 used for data, 1 for parity, advantages and disadvantages are the same as mirroring but duplexing provides much faster read speeds than mirroring.

Striping: (with and without parity) - Data striping is when blocks or bits of data are written to each drive in the array in succession. It's used in most RAID levels and is great for improving read/write speeds because the I/O request are being distributed between all I/O data channels. Parity checking relies on an extra bit called a parity bit, which is used to compare the bit string to an odd or even count. If the odd or even count is not matched based on the setting of the parity bit, then the data string is sent again. Extra drive space is used for the parity bits. Not using parity will improve overall data transmission because of the omission of the parity bit calculation, but should be used when speed is of greater importance than fault tolerance. RAID 5 provides the best fault tolerance because it uses several drives with block interleaving, a distributed check sum for parity and has fast reads.

Volumes: Segments of hard drives which may be contained in one physical drive, span several physical drives, or one drive may contain several segments.

Tape backup: Magnetic tape contains high capacity backup solution at a slow speed. Advantages include the ability to store tapes off site for backup, large capacity for backup, and inexpensive media. A disadvantage is its slow speed.

Prev Page / Next Page: RAID Levels