RAID stands of Redundant Array of Independent Disks. It is used to describe a storage systems' resilience to disk
failure through the use of multiple disks and by the use of data distribution and correction techniques.
RAID can be software, hardware or combination of both.
SOFTWARE RAID uses more system resources as more disk ports and channels are required. It may have lower
cost than hardware RAID because it has no dedicated RAID controller but has lower performance.
HARDWARE RAID offloads parity generation and checking. It also allows for greater disk capacity per disk port.
It requires expensive RAID controller.
RAID has got many levels.
Level 0: Also known as disk stripping, because it uses a disk file system called a strip set. This level does not
provide fault tolerance. Data is divided into blocks and is spread in a fixed order amond all the disks in the array.
This level improves read and write performance by spreading operations across multiple disks, so that operations
can be performed independently.
Level1: This level is also known as disk mirrroring because it uses a disk file system called a mirror set. This level
provides fault tolerance.
It provides a redundant, identical copy of a selected disk. All data written to the primary disk is written to the
mirror disk. It also generally improves read performance but may degrade write performance.
Level 2: RAID level 2 uses error correction algorithm that employs disk stripping strategy that breaks a file into
bytes and spreads it across multiple disks. The error correction method requires several disks. RAID level 2 is
more advanced than level 0, because it provides fault tolerance, but is not as efficient as other RAID levels and
is hence not generally used.
Level 3: It is similar to level 2, because it uses the same stripping method as level 2, but it requires only one disk
for parity data. This level suffers from a write bottleneck, because all parity data is written to a single drive, but
provides some read and write performance improvement.
Level 4: It is similar to level 3, because the stripping method stands the same and requires only one disk for
parity data, but it employs striped data in much larger blocks or segments. This level is not as efficient as level 5
because(as in level 3) all parity data is written to a single drive, so it also suffers from a write bottleneck.
Level 5: Known as Stripping with parity. It is most popular and is similar to level 4 in that it stripes the data in
large blocks across all the disks in the array. It differs in that it writes tha parity across all the disks. The data
redundancy is provided by the parity information. Data and parity information are arranged on the disk array so
that the two are always on different disks. It also has better performance than level 1 and provides fault tolerance.
Keywords: Disk stripping, mirroring, RAID, RAID levels, Redundant disks, storage, stripping with parity
