Table of Contents
Advertisement
RAID 3
RAID 3 provides disk striping and complete data redundancy though a dedicated parity drive. RAID 3 breaks up data
into smaller blocks, calculates parity by performing an exclusive-or on the blocks, and then writes the blocks to all
but one drive in the array. The parity data created during the exclusive-or is then written to the last drive in the
array. If a single drive fails, data is still available by computing the exclusive-or of the contents corresponding strips
of the surviving member disk. RAID 3 is best for applications that require very fast data- transfer rates or long data
blocks.
RAID 5
RAID 5 is sometimes called striping with parity at byte level. In RAID 5, the parity information is written to all of the
drives in the controllers rather than being concentrated on a dedicated parity disk. If one drive in the system fails,
the parity information can be used to reconstruct the data from that drive. All drives in the array system can be used
for seek operations at the same time, greatly increasing the performance of the RAID system. This relieves the write
bottleneck that characterizes RAID 4, and is the primary reason that RAID 5 is more often implemented in RAID
arrays.
RAID 6
RAID 6 provides the highest reliability. It is similar to RAID 5, but it performs two different parity computations or
the same computation on overlapping subsets of the data. RAID 6 can offer fault tolerance greater than RAID 1 or
RAID 5 but only consumes the capacity of 2 disk drives for distributed parity data. RAID 6 is an extension of RAID 5
but uses a second, independent distributed parity scheme. Data is striped on a block level across a set of drives, and
then a second set of parity is calculated and written across all of the drives.
123
Advertisement
Table of Contents
