A logical disk created from available space and
made up of one or more segments on one or more physical disks. Arrays
are typically used to provide data redundancy or enhanced I/O performance.
Background initialization of a redundant array.
The array is accessible throughout. RAID 1 copies the contents of
the primary drive to a secondary drive.
Fast-access memory on the controller that serves
as intermediate storage for data that is read from, or written to, drives.
capacity
Total usable space available in megabytes or gigabytes.
channel
Any path, or bus, used for the transfer of
data and the control of information between storage devices and a RAID
controller.
D
degraded
A redundant (for example, a RAID 1) array in
which one or more members have failed. The data is intact but redundancy
has been compromised. Any further failure would cause the array to fail
and result in data loss.
discovery
The process by which information about a system
or subsystem is obtained.
Consists of the bus number, SCSI ID, and LUN, for
example, 1:04:0. See also bus; LUN; SCSI ID.
drive light
An indicator light on a disk that illuminates during
read or write operations.
E
enclosure
A physical housing for hard disks, which can be
connected internally or externally to a computer. An enclosure usually
contains one or more power supplies, fans, and temperature sensors.
enclosure ID
Consists of the bus number, SCSI ID, and LUN, for
example, 1:04:0. See also bus; LUN; SCSI ID.
A notification or alert from the system indicating
that a change has occurred.
event log
A file used to maintain information about prior
controller activities or errors.
event notification
The means used by the system to communicate information
about event occurrence.
F
failed
State of a nonredundant array that has suffered
a single drive failure, or a redundant array that has suffered multiple
drive failures. A failed array is inaccessible and data is lost.
The ability of a system to keep working in the event
of hardware or software faults. Fault tolerance is usually achieved by
duplicating key components of the system.
fault-tolerant arrays
Arrays that have duplicate components (RAID 1 and
RAID 10).
G
Global Spare
A spare hard disk that automatically replaces a
failed hard disk on any array associated with any HBA.
H
hard disk
A physical hard disk on a SCSI bus.
host adapter
Electronic circuitry that supports I/O to storage
devices, but which does not support RAID.
HostRAID
Intelligent RAID on chip. Adaptec product that adds
entry-level RAID functionality to SCSI ASICs and host bus adapters.
hot spare
A hard drive, an array member, that is not used
in day-to-day data storage but, instead, is reserved for use as a replacement
for one of the other drives in the array in the event of a failure.
I
initialize
Process of preparing a disk for use by the controller.
When a disk is initialized, the controller records the RAID signature
on the disk.
initialized array
An array that is ready for data reads and writes.
L
logical device
A read/write resource, such as an array or spare,
that does not exist until configured by the Software Management tool.
Process performed by the drive firmware that completely
removes the file system from the disk.
LUN
Stands for Logical Unit Number. The number assigned
to a subdevice (logical unit) of a SCSI target. Each SCSI target can contain
up to eight subdevices numbered 0 through 7; however, most SCSI devices
contain only one subdevice (LUN 0).
M
mirroring; mirrored array
Synonym for RAID 1. Full redundancy is obtained
by maintaining identical copies of data on two or more hard disks. Mirroring
provides fault tolerance with a minor performance penalty.
monitoring
The process of receiving, displaying, and logging
system events.
O
off-line array
An array that is not responding.
optimal
The state of an array when it is fully operational.
For redundant arrays, the entire array is protected.
P
parity
In RAID, a form of redundancy used to recreate the
data of a failed hard disk in a disk array. Used in RAID levels 2, 3,
4, and 5.
physical resources
The Adaptec Family Adapter products, channels, and
SCSI devices installed in the server.
poll
A validation of discovery information.
R
RAID
Acronym for Redundant Array of Independent (Inexpensive)
Disks. Coined in 1987 by researchers at the University of California at
Berkeley to describe a set of disk array architectures that provide fault
tolerance and improved performance.
RAID 0 (striping)
A single-level array consisting of two or more equal-sized
segments residing on different disks. RAID 0 distributes data evenly
across its respective drives in equal-sized sections called stripes. RAID 0
arrays are not redundant.
RAID 1 (mirrored)
Single-level array consisting of two equal segments
residing on two different drives. Provides redundancy by storing identical
copies on two drives.
rebuild
The regeneration and writing onto one or more replacement
hard disks of all the data from a failed hard disk in a redundant RAID
array. Rebuild is usually performed in the background while applications
are accessing data in the array.
Maintaining duplicates for any given component in
a system so the system can automatically replace a failed component with
a working substitute. A parity RAID array's members are redundant since
surviving members may collectively replace the data from a failed hard
disk.
S
SCSI
Stands for Small Computer System Interface. A parallel
interface standard used to connect various types of peripheral device,
(such as disk drives) to computers. Although SCSI is an ANSI standard,
it exists in several variations.
SCSI array
A group of hard disks connected to the same SCSI
array controller and accessed as a single storage unit. It is not apparent
to users that an array is actually more than one disk drive.
SCSI array controller
The electronic circuitry used to control a SCSI
array.
The number assigned to each SCSI device attached
to a SCSI bus. See also bus; disk
ID.
server
A computer set up to provide services used by other
computers in a network environment. Typical servers handle disk storage,
e-mail, or database functions.
SMART
Stands for Self-Monitoring Analysis and Reporting
Technology. Hard disks that support this technology continually analyze
their performance and generate an alert if they determine that the hard
disk is likely to fail in the next few hours.
spare
A disk drive that is assigned to one or more arrays
and kept in reserve in case of drive failure. A spare automatically replaces
a failed drive in an array.
stand-alone hard disk
A hard disk that is not part of an array.
stripe size
The size that is used to stripe data or parity information
across the disks in the array.
striping, disk; striped array
Spreading data over multiple disks to enhance performance.
Often referred to as RAID 0, simple data striping provides no redundancy
scheme and is not technically a standardized RAID level.
subsystem
The collection of software and hardware that controls
storage to be managed and monitored.
system
Used generically to refer to a server, a networked
client, or a stand-alone workstation.
An operation that occurs only on the RAID controller,
asynchronous to all other operations; for example, initializing a disk
or verifying an array. Some tasks can take an extended period of time,
particularly if the RAID controller is also performing intensive data
I/O while the task is running.
V
verify
The process of checking the integrity of redundant
data stored on fault-tolerant arrays.
virtual device
Arrays and spares that do not exist until they are
configured on the server with HostRAID.
virtual device order
The sequence in which the server's operating system
detects the arrays, single hard disks, and other devices connected to
the controller when the server boots.
