IBM ServeRAID, Справочник пользователя

Страница: 215 / 226

Glossary
A
array. Two or more physical drives combined for
increased capacity, speed, or fault tolerance. Once
created, an array is configured as one or more logical
drives. The operating system in the server recognizes a
logical drive as a single drive.
auto-synchronization. Synchronization on RAID
level-5, level-5E, and level-50 drives that is
automatically initiated when logical drives are created.
This type of synchronization works in the background.
B
battery-backup cache. Buffer storage that protects
data during write-back operations; in the event of a
power failure, it preserves the data in the controller
cache.
BIOS-compatibility mapping. A ServeRAID controller
setting with two options: Limited and Extended . The
default option is Extended .
BIOS-compatibility mapping—Extended. An option
of BIOS compatibility mapping. When the BIOS
compatibility mapping is set to
Extended , the
ServeRAID BIOS is aware that the operating system
supports 8 GB or smaller physical drives.
BIOS-compatibility mapping—Limited. An option of
BIOS compatibility mapping. When the BIOS
compatibility mapping is set to
Limited , the ServeRAID
BIOS is aware that the operating system supports 2 GB
or smaller physical drives. This enables the user to
migrate data from older PCI or Micro Channel
®
adapters.
block. A data unit created when data is striped across
physical drives.
blocked. A logical-drive state in which the data in a
RAID level-0 logical drive might be damaged and will
need to be restored. After an array fails and a rebuild
operation is initiated, the ServeRAID controller sets
RAID level-0 logical drives to the blocked state.
C
cluster. In ServeRAID technology, two independent
computer systems organized into a network for the
purpose of sharing resources and communicating with
each other. A cluster provides fault-tolerance at the
server level.
compression. The process by which a RAID level-5E
logical drive utilizes the distributed spare drive in the
event of a physical drive failure. After the data is
reconstructed, the original array undergoes compression
and the distributed spare drive becomes part of the new
array.
controller. A device that coordinates and controls the
operation of one or more input/output devices, such as
workstations, and synchronizes the operation of such
devices with the operation of the system as a whole.
controller names. In a cluster or failover pair, the
unique names that identify the members of the pair to
one another.
critical. The state of a RAID level-1, level-1E, level-5,
level-5E, level-10, level-1E0, or level-50 logical drive
that contains a defunct drive.
D
data mirroring. A technique that creates a single
logical drive from two physical drives. All data written to
the logical drive is written to both physical drives,
creating a pair of physical drives containing exactly the
same data.
data scrubbing. A feature that provides automatic,
continuous synchronization during system use. This
features works in the background, and ensures that the
redundant data and/or parity is correct.
data striping. A technique that divides a logical drive
into data blocks, called stripes, which are then
distributed over the physical drives. The layout is such
that a sequential read of data on the logical drive results
in parallel reads to each of the physical drives, resulting
in improved performance.
decompression. The process by which a compressed
array returns to its original striping scheme, after a
failed drive is replaced in a RAID level-5E logical drive.
Contrast with compression.
defunct. A physical-drive state in which the
ServeRAID controller cannot communicate properly with
the drive.
distributed spare drive. In RAID level-5E, the logical
drive designated as a spare drive. Because this spare
drive is spread over several physical drives, it is
described as distributed.
E
enclosure. A storage expansion unit, such as the
Netfinity
®
EXP300, that contains SCSI disk drives.
© Copyright IBM Corp. 2002
199