IBM ServeRAID, User Reference

Page: 217 / 226

parity block. In a RAID level-5, level-5E, or level-50
logical drive, a data unit that contains a representation
of the data from other blocks in the same stripe.
physical drive. A hard disk drive.
POST. Power-on self-test. During POST, the
ServeRAID controller compares the stored configuration
information to the configuration that is actually present.
If a discrepancy exists, one or more status messages
appear after the POST completes but before the
operating system loads.
Q
quorum drive. In the Microsoft Clustering Solution, a
logical drive that stores the quorum resource
information.
R
rack enclosure. See enclosure.
RAID. A technology of grouping several physical drives
in a computer into an array that you can define as one
or more logical drives. Each logical drive appears to the
operating system as a single drive. This grouping
technique greatly enhances logical-drive capacity and
performance beyond the physical limitations of a single
physical drive.
RAID level-0. A RAID level that uses data striping to
distribute data evenly across physical drives. While it
enables full utilization of physical drive capacity and
performance acceleration, RAID level-0 provides neither
fault tolerance nor redundancy.
RAID level-1. A RAID level that uses data mirroring to
distribute data across two physical drives. It provides
data redundancy and performance acceleration,
although the usable physical drive space is reduced by
50 percent.
RAID level-1E. A RAID level that uses both data
striping and data mirroring to distribute data across
three or more physical drives. Data is striped across
each disk in the array; the first set of stripes are the
data stripes, and the second sets of stripes are mirror
copies of the first stripe, shifted one drive. It provides
data redundancy and performance acceleration,
although the usable physical drive space is reduced by
50 percent.
RAID level-5. A RAID level that uses data striping and
block interweaving to distribute data across three or
more physical drives. It provides full data protection and
performance acceleration, although only 67-94% of
physical drive storage capacity can be used.
RAID level-5E. A RAID level that uses data striping
and block interweaving to distribute data across four or
more physical drives; it uses some space on each
physical drive as a distributed hot-spare. It provides full
data protection and performance acceleration, although
only 50-88% of physical drive storage capacity can be
used.
RAID level-x0. Refers to RAID level-00, level-10,
level-1E0, and level-50. These RAID levels use
spanned arrays (arrays of arrays) to enable the use of
up to 60 physical drives. RAID level-00 provides no data
redundancy, but provides performance acceleration and
allows 100% of the physical drive storage capacity to be
used. RAID levels -10, -1E, and -50 provide full data
protection, performance acceleration, and greater
reliability, although only 67-94% of physical drive
storage capacity can be used.
read-ahead cache mode. A ServeRAID controller
setting that determines whether the ServeRAID
controller transfers data from disk to its local cache in
increments equal to the stripe-unit size or the system
I/O requests. The options are
enabled , disabled , and
adaptive ; the default option is adaptive.
read-ahead cache mode—Adaptive. The
default-value of the read-ahead cache mode. When the
read-ahead cache mode is set to adaptive, the
ServeRAID controller continually reevaluates whether to
set the read-ahead cache mode to
enabled or
disabled .
read-ahead cache mode—Disabled. An option of the
read-ahead cache mode. When the read-ahead cache
mode is disabled, the ServeRAID controller transfers
data from the logical drive to its local cache in
increments equal to the system I/O request size. This
optimizes performance when the workload is random or
the system I/O requests are smaller than the stripe-unit
size.
read-ahead cache mode—Enabled. An option of the
read-ahead cache mode. When the read-ahead cache
mode is enabled, the ServeRAID controller transfers
data from the logical drive to its local cache in
increments equal to the stripe-unit size. This optimizes
performance when workloads are steady and
sequential.
ready. A physical-drive state in which the drive is
available for definition.
rebuild. An operation to reconstruct data after the
problem that caused a physical drive to become defunct
has been resolved.
rebuilding. The state of a physical drive undergoing a
rebuild operation.
rebuild rate. The speed (high, medium, or low) at
which a rebuild operation will occur.
redundant array of independent disks (RAID). See
RAID.
Glossary
201