| Qlogic |
A brand
of host bus adapters and switches. |
| QoS
|
Quality
of Service. |
| RAID |
Redundant
Array of Inexpensive Disks: when configured for performance a RAID writes
and reads data in parallel from multiple drive simultaneously. In theory
data can be moved at the speed of one drive multiplied by the number of
drives working in parallel, although in practice management overheads reduce
this significantly. |
| RAM |
Random
Access Memory: the PC's primary storage area, used to write, store and retrieve
information and program instructions which are then passed to the CPU for
processing. The type of RAM used affects performance as the information
stored here has to be refreshed many times per second by the processor.
Manufacturers are continually coming up with new designs to provide the
fastest possible access times at the lowest possible cost. |
| Redundancy |
Having
multiple occurrences of a component to maintain high availability. |
| Reed-Solomon
ECC |
Reed-Solomon
error correction codes are special and widely implemented because they are
"almost perfect" in the sense that the extra (redundant) letters
added on by the encoder is at a minimum for any level of error correction,
so that no bits are wasted. It also allows correction of "erasures"
which is like taking a pencil eraser and erasing a letter in a word. The
letter that should be in that position is unknown, but the position of the
erasure is known. With Reed-Solomon error correction, you get more correction
power per dollar by being able to correct multiple randomly positioned bytes
in error. |
| Refresh |
The process
used to restore the charge in DRAM memory cells at specified intervals.
The required refresh interval is a function of the memory cell design and
the semiconductor technology used to manufacture the memory device. There
are several refresh schemes that may be used. |
| SAN |
Storage
Area Network. |
| SCA |
Single
Connector Attachment: Same speed SCSI interface as LVD, but integrates power
and I/O information into a single 80-pin connector. Used in high-end servers
to allow hard disks to be hot-swapped in a RAID array. |
| Scalability |
The capacity
of hardware, software and networks to change size according to the number
of users that they accommodate. Most often, scalability refers to the capacity
to expand rather than shrink. |
| SCSI |
Small
Computer Systems Interface is the standard used in many computers to communicate
with peripherals like scanners and modems. SCSI adheres to an ANSI standard,
and SCSI models have independent controller logic. |
| SCSI-2 |
An updated
version of the SCSI bus architecture. SCSI-2 describes the latest published
ANSI standard (X3.131-1994) that prescribes several connectors (both shielded
and unshielded) that include 1 byte wide data bus, defines FAST transfer
speeds, defines SCSI protocol for wider data transfers, defines the parallel
SCSI messages, and command structure. This provides the base on which future
SCSI features are compared against. These enhancements, which include caching,
command queuing and power management, increased its performance and flexibility. |
| SCSI-3 |
This term
describes a set of related standards that are currently being developed.
The SCSI-3 standards are layered to allow substitution of parts of the structure
as new technology emerges. For example, a comparable set of standards for
a SCSI Fiber Channel interface disc drive replaces the physical and protocol
layers with new documents but uses the same documents for the other 3 "layers".
The basic layers are: physical (connectors, pin assignments, electrical
specifications), protocol (description of how physical layer activity is
organized into bus phases, packets, etc.), architecture (a description of
how command requests are organized, queued, responded to by any protocol),
primary commands (description of commands that must be supported by all
SCSI devices), and device specific commands (commands that are specific
to a particular class of devices; CD-ROMs or WORM drives, for example). |
| SCSI
Fast |
This refers
to timings defined in SCSI-2 for 10 MegaTransfer/sec transfer rate. A "MegaTransfer"
refers to the rate of signals on the interface regardless of the width of
the bus. For example, 10 MT/sec rate on 1 byte wide bus results in 10 MB/sec
transfer rate but on a 2 byte wide bus results in a 20 MB/sec transfer rate. |
| SCSI
Fast-20 |
This refers
to timings defined in SCSI-3 physical document for 20 MT/sec transfer rate.
This achieves data rates twice as fast as SCSI FAST rates. |
| SCSI
Fast-40 |
This refers
to timings being defined for a future revision of the SCSI-3 physical documents
that achieves 40 MT/sec. |
| SCSI
Wide |
This term
usually refers to the two byte wide (68 pin) connector that is defined in
the SCSI-3 Parallel Interface (SPI) document. This technically makes it
a SCSI-3 feature. The term can be generically applied to any implementation
wider than 1 byte, but there are no implementations wider than 2 bytes today.
Wider implementations aren't expected because faster transfer rates are
giving plenty of life to 2 byte transfers until serial interfaces (like
Fibre Channel or FireWire) become more popular. |
| SCSI
Fast Wide |
This refers
to a combination of SCSI Fast transfer rate with 2 byte wide connector,
which results in 20 MB/sec data transfer rate. |
| Serial |
The transmission
of data bits in sequential order over a single line. |
| Server |
A computer
that processes end-user applications or requests. |
| SES |
SCSI Enclosure
Services. |
| SFF |
Small
Form Factor. |
| SFP |
Small Form
Factor Pluggable. |
| Shock
Rating |
A rating
(expressed in Gs) of how much shock a disk drive can sustain without damage.
Operating and non-operating shock levels are usually specified separately. |
| Single
mode |
A fiber
optic cabling specification that provides up to 10km and > 50km distances
between devices. |
| SMART
(Self-Monitoring Analysis and Reporting Technology) |
SMART monitors
the disk drive as it is working and warns the user of possible drive failures
by checking key device performance indicators and reporting significant
changes above the recommended level. This gives the user an opportunity
to backup the files on the hard drive. |
| SMF |
Single
mode fiber. |
| SMI |
Structure
of Management Information. |
| SoIP |
SCSI over
IP. |
| Soft
Error |
A faulty
data reading that does not recur if the same data is reread from the disk
or corrected by ECC. Usually caused by power fluctuations or noise spikes. |
| Software
RAID |
Uses the
server processor to perform RAID calculations. Host CPU cycles that read
and write data from and to disk are taken away from applications. Software
RAID is less costly than dedicated hardware RAID storage processors, but
its data protection is less efficient and reliable. |
| SONET |
Synchronous
Optical Network. |
| Spindle |
The drive's
center shaft, on which the hard disks are mounted. A synchronized spindle
is a shaft that allows two disks to spin simultaneously as a mirror image
of each other, permitting redundant storage of data. |
| SPOF |
Single
Point of Failure. |
| SRAM |
Static
Random Access Memory: a form of RAM that retains its data without the constant
refreshing that DRAM requires. SRAM is generally preferable to DRAM because
it offers faster memory access times, but it is also more expensive to manufacture
because it contains more electrical components. |
| SSA |
Serial
Storage Architecture: a peripheral interface from IBM whose ring configuration
allows remaining devices to function if one fails. SCSI software can be
mapped over SSA allowing existing SCSI devices to be used. |
| SSD |
Solid State
Disk storage. |
| Storage |
A device
used to store data; such as a disk or tape. |
| Store-and-forward |
A switching
technique that requires buffering an entire frame before a routing decision
is made. |
| Striping |
A RAID
technique for writing a file to multiple disks on a block-by-block basis,
with or without parity. |
| Sustained
Transfer Rate |
The amount
of data a drive can continuously read or write per second |
| Switch |
A Fabric
device providing bandwidth and high-speed routing of data via link-level
addressing. |
| Synchronous |
Refers
to events that are synchronized, or coordinated, in time. Communication
within a computer is usually synchronous and is governed by the microprocessor
clock. Signals along the bus, for example, can occur only at specific points
in the clock cycle. |
| T10 |
A
standards committee chartered with creating standards for SCSI. |
| T11 |
A standards
committee chartered with creating standards for Fibre Channel. |
| T13 |
A standards
committee responsible for all interface standards relating to the popular
AT Attachment (ATA) storage interface utilized as the disk drive interface
on most personal and mobile computers today. |
| Target |
A
disk or a tape device in SCSI. |
| TCP/IP |
Transmission
Control Protocol over Internet Protocol. |
| Topology |
The physical
or logical arrangement of devices in a networked configuration. |
| Transceiver |
A device
that converts one form of signaling to another for transmission and reception;
in fiber optics it means optical to electrical. |
| Transfer
Rate |
The rate
at which the disk sends and receives data from the controller. The sustained
transfer rate includes the time required for system processing, head switches
and seeks, and accurately reflects the drive's true performance. The burst
mode transfer rate is a much higher figure that refers only to the movement
of data directly into RAM. |
