IBM ServeRAID, Справочник пользователя

Страница: 218 / 226

restore to factory-default settings. An action that
restores all parameters in the controller to the
factory-default settings. If logical drives are defined, the
data stored on them will be lost.
S
SCSI. See small computer system interface.
SCSI ID. A unique ID assigned to each SCSI device
connected to a SCSI controller. This ID enables the
controller to identify the device and ensure that different
devices on the same SCSI channel do not transfer data
simultaneously.
SCSI initiator ID. In a failover or clustering
environment, the ID assigned to a SCSI controller. Each
ServeRAID controller and its partner must have different
SCSI initiator IDS; one must be 6 and the other 7.
SCSI transfer speed. The speed at which data can be
transferred between a physical drive and the ServeRAID
controller.
ServeRAID Manager. A program used to configure
ServeRAID controllers, view the ServeRAID
configuration, create arrays and logical drives, delete
arrays, dynamically increase the logical-drive size,
change RAID levels, and more.
ServeRAID Mini-Configuration program. A program
that allows the user to display the ServeRAID controller
settings, and to perform a limited set of configuration
functions without using the
IBM ServeRAID Support CD.
ServeRAID ROM Update Wizard. A program that
updates the BIOS and firmware codes on ServeRAID
controllers.
shared drives. The physical drives controlled by a
cluster or failover pair.
small computer system interface. A standard
hardware interface that enables a variety of peripheral
devices to communicate with one another.
spanned array. An array of arrays. Used in RAID
level-00, level-10, level-1E0, and level-50 to permit the
use of larger numbers of physical drives. The spanned
array contains arrays, each of which contains a
sub-logical drive, which can be RAID level-0, level-1,
level-1E, or level-5. The RAID-level for the logical drive
contained within the spanned array is 0.
standby hot spare. A hot-spare physical drive that the
ServeRAID controller has spun down. If an online drive
becomes defunct and no suitable hot-spare drive is
available, a standby drive of the appropriate size
automatically spins up and enters the rebuild state.
stripe-unit size. The granularity at which data is
stored on one drive of the array before subsequent data
is stored on the next drive of the array. The
performance of a ServeRAID controller is maximized if
the stripe-unit size is close to the size of the system
input/output requests.
stripes. The collection of stripe units, from the first to
last drive of the array.
sub-logical drive. In a RAID level-x0 configuration, a
logical drive contained within each array of the spanned
array. A sub-logical drive can be RAID level-0, level-1,
level-1E, or level-5.
synchronization. The process of recalculating and
rewriting either redundant data (RAID level-1, level-1E,
level-10, and level-1E0 logical drives) or parity (RAID
level-5, level-5E, and level-50 logical drives).
T
throughput. The speed at which data can be moved
from one place to another, usually expressed in MB per
second.
U
unattended mode. A ServeRAID controller setting that
determines how the BIOS handles failures during
system startup. The options are
enabled and disabled ;
the default value is disabled .
unattended mode—Disabled. An option of the
unattended mode. When the unattended mode is
disabled and the BIOS detects a failure, the system
stops at the recovery option screen and waits for the
user to respond.
unattended mode—Enabled. An option of the
unattended mode. When the unattended mode is
enabled and the BIOS detects a failure, the system
waits for 30 seconds for the user to respond to the
recovery option screen. Then, the BIOS automatically
selects an appropriate option and continues the system
startup process. Unattended mode must be enabled in a
clustering environment.
W
write-cache mode. A ServeRAID controller setting that
determines whether the controller writes data to the
physical drive before or after sending a confirmation to
the operating system. The settings are
write back and
write through .
write-cache mode—write back. A setting of the
write-cache mode. When the write-cache mode is set to
write back and the operating system sends data to the
controller, the controller sends a confirmation back to
the operating system before actually writing the data to
a storage device. This increases performance, but, if a
battery-backup cache is not used, increases the risk of
data loss in the event of a power failure.
202
IBM ServeRAID: User’s Reference, Version 5.10