Contemporary disc systems of RAID
RAID-3.
A segment is physically located on the first drive, V-segment on the second disc
in the same place as X segment is placed on the third disc, segment 0-following
the A segment on the first drive, the E-segment in the second segment, followed
by B and so on until as long as the size of the entire drive will not be
distributed. Redundancy provides a fourth disc. The first segment of the fourth
disc contains an amount modulo 2 (Hog) information segments A, B and C, the
first three discs. The next segment of the fourth disc contains Xor (D, E, F),
etc. When reading data segment is defined physical address on the disk
containing the data. Then read data from all disks simultaneously. When
recording also made a request for the display of physical addresses, and
recordable data simultaneously posted on all drives. In the process of recording
Xor value is calculated for all traffic, and the result has also written to the
disk simultaneously. With disk failure data on it can be restored to data from
the healthy Hog-diska and the remaining disks. RAID-3 differs from RAID-1 so
that the segment size is selected taking into account the mechanism of
redundancy. During the operations, or read-write with a crashed disk requires
that vssegda Hog-segmentov calculation was possible. For this segment, the size
should be small compared to the size of the request that all segments involved
in the operation were available in the Hog each time. The typical size segment
in RAID-3 a bit or a byte. The request, distributed to all disks, means that
every disc will use its portion of the information on request at the same time
as drutimi. This ensures a very high speed exchange for large requests, but
limited by the number of requests serviced at the same time, because every time
you can serve only one request. This provides good performance for load, which
is a consistent flow of high demand. RAID-3 provides a good data integrity,
since the loss of data is possible only if down another drive will be available
before the first refused to be replaced. RAID-2 is very similar to RAID- Z, but
instead of a surplus Hog-diska he uses several, which is formed Hamming code and
redundant storage inforatsii require some additional disks instead of one.
Redundancy can be more expensive.
Next part
|