The section below provides an overview of the most common terminologies related to motherboard industry.

| A | B | C | D | E | F | G | H | I | J | K | L | M |
N | O | P | Q | R | S | T | U | V | W | X | Y | Z |

Adapter Card - A circuit board that plugs into a computer's expansion bus and gives the computer additional capabilities.

Audio Codec '97 (AC '97) defines a high-quality, 16-bit audio architecture for the PC that is used in the majority of today's desktop platforms. [top]

ACPI (A dvanced Configuration and Power Interface) is a power management specification that allows the operating system to control the amount of power distributed to the computer's devices.  Devices not in use can be turned off, reducing unnecessary power expenditure.  ACPI defines a new interface to the system board, and enables the OnNow design initiative for instantly available PCs. [top]

AGP (Accelerated Graphics Port) is a PCI-based interface that was designed specifically for demands of 3D graphics applications.  The 32-bit AGP channel directly links the graphics controller to the main memory.  While the channel runs at only 66 MHz, it supports data transmission during both the rising and falling ends of the clock cycle, yielding an effective speed of 133 MHz. [top]

AGP Pro is an extension to the AGP interface specification that meets the needs of advanced workstation graphics. This specification is primarily designed to deliver additional electrical power to the graphics add-in cards. The AGP Pro definition includes an extended connector, thermal envelope, and mechanical specifications for cards, I/O brackets, and motherboard layout requirements. [top]

AIMM (AGP In-line Memory M odule) - 4 MB Display Cache card that plugs into the AGP port to gain additional performance. [top]

AMR (Audio Modem Riser) is Intel's specification on motherboard design.  Motherboard with this type of architecture allows for designs w/o analog I/O functions.  However, these functions can be added by the codec chip on a riser card, which plugs into the motherboard perpendicularly, resulting in better audio quality. [top]

Array is two or more hard disk drives grouped together to appear as a single device to the host computer. [top]

AT was the original form factor of IBM's PC. [top]

ATAPI (AT Attachment Packet Interface), also known as IDE or ATA, is a drive implementation that includes the disk controller on the device itself.  It allows CD-ROMs and tape drives to be configured as master or slave devices, just like hard drives. [top]

ATX form factor was designed to replace the AT form factor.  It improves on the AT design by rotating the board ninety degrees, so that the IDE connectors are closer to the drive bays, and the CPU is closer to the power supply and cooling fan.  The keyboard, mouse, serial, USB, and parallel ports are built in. [ top]

Bandwidth refers to carrying capacity.  The greater the bandwidth, the more data the bus, phone line, or other electrical path, can carry.  Greater bandwidth, then, also results in greater speed. [top]

BBS (BIOS Boot Specification) is a feature within the BIOS that creates, prioritizes, and maintains a list of all Initial Program Load (IPL) devices, and then stores that list in NVRAM.  IPL devices have the ability to load and execute an OS, as well as provide the ability to return to the BIOS if the OS load process fails for some reason.  At that point, the next IPL device is called upon to attempt loading of the OS. [ top]

The BIOS (Basic Input/Output System) program resides in the ROM chip, and provides the basic instructions for controlling your computer's hardware.  Both the operating system and application software use BIOS routines to ensure compatibility. [top ]

Bootable Array Support refers to the ability to make the system boot from a RAID array instead of from a standalone (single) disk. [ top]

BMC Commonly integrated into server system boards that support IPMI, a Baseboard Management Controller (BMC) normally features embedded functions, multiple interfaces, and a generous number of general purpose I/O pins. Other common features associated with a BMC would be a complete IPMI protocol stack with IPMB support for intra-chassis communications, as well as ICMB for inter-chassis communications. [ top]

Boot Disk - A floppy disk that contains information necessary to start a computer when the hard drive is not functioning correctly. [top ]

Bps (Bits per Second) - The standard measure of data transmission speeds. [top]

A buffer is a portion of RAM that is used to temporarily store data, usually from an application, though it is also used when printing, and in most keyboard drivers.  The CPU can manipulate data in a buffer before copying it, all at once, to a disk drive.  While this improves system performance -- reading to or writing from a disk drive a single time is much faster than doing so repeatedly -- there is the possibility of losing your data should the system crash.  Information stored in a buffer is temporarily stored, not permanently saved. [ top]

A bus is a data pathway.  The term is used especially to refer to the connection between the processor and system memory, and between the processor and PCI or ISA local buses.

Bus Mastering allows peripheral devices and IDEs to access the system memory without going through the CPU (similar to DMA channels). [top ]

A cache is a temporary, fast storage area that holds data from a slower storage device for quick access as needed by an application.  Access time is fast using a cache, because the needed information is stored in the SRAM instead of in the slower DRAM.  Note that the cache is also much smaller than your regular memory: a typical cache size is 512KB, while you may have as much as 2GB of regular memory. [ top]

Cache size refers to the physical size of the cache onboard.  This should not be confused with the cacheable area, which is the total amount of memory that can be scanned by the system in search of data to put into the cache.  A typical setup would be a cache size of 512KB, and a cacheable area of 512MB.  In this case, up to 512MB of the main memory onboard is capable of being cached.  However, only 512KB of this memory will be in the cache at any given moment.  Any main memory above 512MB could never be cached. [ top]

Clock - An electronic circuit that generates evenly spaced pulses at speeds of millions of cycles per second.  The pulses are used to synchronize the flow of information through the computers internal communication channels. [ top]

Clock Speed - The speed of the internal clock of a microprocessor that sets the pace-measured in megahertz (MHz) at which operations proceed within the computers internal processing circuitry. [ top]

Closed and open jumpers  Jumpers and jumper pins are active when they are On or Closed, and inactive when they are Off or Open. [top]

CMOS (C omplementary Metal-Oxide Semiconductors) are chips that hold the basic start-up information for the BIOS. [top ]

CNR (Communication & Network Riser) [top]

Codec (Coder/decoder) is a filter that manipulates data in some form, usually by compressing or decompressing the data stream. [top]

Cold Boot - A system start up initiated by turning on the systems power switch. [top]

The COM port is another name for the serial port, which is so-called because it transmits the eight bits of a byte of data along one wire, and receives data on another single wire (that is, the data is transmitted in serial form, one bit after another).  Parallel ports transmit the bits of a byte on eight different wires at the same time (that is, in parallel form, eight bits at the same time). [ top]

Co-processor resides on an array that relieves the host CPU from executing processor-intensive operations such as RAID 5 parity calculations and secondary RAID 1 writes. [top]

CPU (Central Processing Unit) is the brain of your computer. It is made up of two parts: The Arithmetic Logic Unit (which does all the processing) and the Control Unit (which makes sure every part of the computer is working together to present the information). [ top]

DC (Display Cache) [top]

DDR (Double Data Rate) is a technology designed to double the clock speed of the memory.  It activates output on both the rising and falling edge of the system clock rather than on just the rising edge, potentially doubling output. [top]

Device Driver - A program that extends the operating systems (OS) capabilities by enabling the operating system to work with a specific hardware device. [top]

DIMM (Dual In-line Memory Modules) - A 64-bit wide, 168-pin module used in Pentium and newer PC's.  They are available in several versions including 5 volt or 3 volt; buffered or unbuffered; with FPM/EDO or SDRAM memory; in 64-bit (non-ECC/parity) or 72-bit (ECC/parity) forms.  Most Pentium and newer PC's require 3.3 volt unbuffered SDRAM DIMMs in either non-ECC or ECC versions. [top]

DIMM banks are sometimes called DIMM sockets, because the physical slot and the logical unit are the same.  That is, one DIMM module fits into one DIMM socket, which is capable of acting as a memory bank. [top]

DMA  Direct Memory Access channels are similar to IRQs.  DMA channels allow hardware devices (like sound cards or keyboards) to access the main memory without involving the CPU.  This frees up CPU resources for other tasks.  As with IRQs, it is vital that you do not double up devices on a single line.  Plug and Play devices will take care of this for you.

In Doze mode, only the CPU's speed is slowed. [top]

DRAM (Dynamic Random Access Memory) - The most common type of computer memory.  DRAM can be made very inexpensively compared to other types of memory.  DRAM chips are small and inexpensive because they normally require only one transistor and a capacitor to represent each bit.  The capacitors must be energized every 15ms or so (hundreds of times per second) to maintain their charges.  DRAM is volatile, meaning it will lose data with no power or without regular refresh cycles. [ top]

Duplexing means mirroring across two RAID cards. [ top]

Dynamic Array Expansion - see "Online Capacity Expansion"

ECC (Error- Correcting Code) functions to test the accuracy of data transmission, both in and out of memory. [top]

EDO DRAM (Extended Data Output DRAM) a faster type of DRAM in that it can start working on the next block of memory at the same time it sends previous one to CPU. [top]

EEPROM  (Electrically E rasable Programmable ROM) also called Flash BIOS, is a ROM chip that can, unlike normal ROM, be updated.  This allows you to keep up with changes in the BIOS programs without having to buy a new chip.  TYAN's BIOS updates. [top]

EPI (Enhanced Parallel Interface) Technical Report describes how to design SCSI systems. The cable lengths and loads are defined by the electrical parameters, allowing system designs to take advantage of faster cables and reduce loads. The technical report describes how to work with wide (68 Pin) and narrow (50 pin) devices on the same system. EPI includes documentation of Expanders, Bridging expanders, Switches and common connectors not documented in the standards. [top]

EPS12V A non-redundant power supply aimed at SSI-based server boards. The specification defines a 450W or 550W power supply with 6 outputs; 3.3v, 5v, 12v1, 12v2, 12v3, -12v, and 5vSB. The form factor is stretch PS/2, with connector/cable assemblies required for the motherboard power, remote sensing and control functions, and peripheral power. [ top]

ESCD (Extended System C onfiguration Data) is a format for storing information about Plug and Play devices in the system BIOS.  This information helps properly configure the system each time it boots. [top]

Expansion card - An integrated circuit card that plugs into an expansion slot on a motherboard to provide access to additional peripherals or features not built into the motherboard.  Also referred to as an add-in card. [top]

Expansion slot - A slot on the motherboard that physically and electrically connects an expansion card to the motherboard and the system busses. [ top]

Fast-20 commonly known as Ultra SCSI is a speed doubling from the SPI document allowing Fast-20, 20 megatransfers per second or on a wide bus 40 megabytes per second. This is not complete standard, it only has the sections required for the Fast-20, all other sections use the SPI standard. [ top]

FAT32 - A disk file allocation system from Microsoft that uses 32-bit values for FAT entries instead of 16-bit values used by the original FAT system, enabling partition sizes up to 2TB (terabytes).  FAT32 first appeared in Windows 95b and is also found in Windows 98 and Windows NT 5.0

Fault Tolerance refers to a system where one component can quickly be replaced without causing a loss of service, such as in a RAID system. [top]

FDISK - The name of the disk partitioning program under several different operating systems (OS) to create the master boot record and all allocate partitions for the OS's use.

Firewire (IEEE1394) - A serial I/O interface standard that is extremely fast, with data transfer rates up to 400MB/sec, 800MB/sec or 3.2GB/sec depending on the version of the standard used.

Firmware is low level software that controls the system hardware. [top]

Form factor is an industry term for the size, shape, power supply type, and external connector type of the PCB (personal computer board) or motherboard.  The standard form factors are the AT and ATX, although TYAN also makes some Baby-AT boards. [ top]

FWH The FirmWare Hub is a key component of the Intel Accelerated Hub Architecture. Able to contain both the system BIOS and integrated graphics/video BIOS on one component, the FirmWare Hub connects directly to the I/O Controller Hub (ICH) without requiring an ISA bus. [ top]

A Global timer is an onboard hardware timer, such as the Real Time Clock. [top]

GUI (Graphical User Interface) - A type of program interface that allows users to choose commands and functions by pointing to a graphical icon using either a keyboard or a pointing device such as a mouse. [top]

Handshaking is a form of encryption.  One system, typically the server, sends an encryption scheme to another agent, typically a client.  Thus, the client's data is protected during transmittal to the server. [top]

HCT (Hardware Compatibility Test) is a suite of tests from WHQL that verifies hardware and device driver operations under a specific operating environment. [top ]

HDD stands for Hard Disk Drive. [top]

Hot Spare - A spare hard drive will automatically be used to replace the failed member of a redundant disk array. [top]

Hot Swap refers to the ability to remove a failed member of a redundant disk array and replace it with a good drive without bringing down the server or interrupting transactions that involve other devices. [top]

H-SYNC controls the horizontal properties of the monitor. [top]

IC (Integrated Circuit) is the formal name for the computer chip. [top ]

ICH The I/O Controller Hub is a highly integrated multifunctional controller that provides the interface to the PCI bus, and integrates many of the functions needed in today's PC platforms. It normally communicates with the host controller (MCH) over a dedicated hub interface. [ top]

IDE (Integrated Device/Drive E lectronics) is a simple, self-contained hard drive interface.  It can handle drives up to 8.4GB in size.  Almost all IDEs sold now are in fact Enhanced IDEs (EIDEs). [ top]

IDE INT (IDE Interrupt) is a hardware interrupt signal that goes to the IDE. [ top]

I/O (Input / Output) is the connection between your computer and another piece of hardware (mouse, keyboard, etc.). [top]

IPL (Initial Program Load): a feature built into BBS-compliant devices, describing those devices as capable of loading and executing an OS, as well as being able to provide control back to the BIOS if the loading attempt fails. [top]

IPMI The Intelligent Platform Management Interface (IPMI) is a hardware level interface specification that defines a common, abstracted, message-based interface to platform monitoring and control functions. As a hardware-level interface, it sits at the bottom of a typical management software stack. Thus, IPMI is "management software neutral." It can be exposed through any standard management software interface, such as WMI, CIM, SNMP or DMI. [ top]

IRQ (Interrupt Request) - Physical connections between external hardware devices and the interrupt controllers.  When a device such as a floppy controller or a printer needs the attention of the CPU, a IRQ line is used to get the attention of the system to perform a task.  On PC and XT IBM compatible systems, 8 IRQ lines are included, numbered IRQ0 through IRQ7.  On the AT and PS/2 systems, 16 IRQ lines are numbered IRQ0 through IRQ15.  IRQ lines must be used by only a single adapter in the ISA bus systems, but Micro Channel Architecture (MCA) adapters can share interrupts. [top]

ISA (Industry Standard Architecture) - The bus architecture that was introduced as an 8-bit bus with the original IBM PC in 1981 and later expanded to 16-bit with the IBM PC/AT in 1984.  ISA slots are still found in many PC systems today. [top]

LAN (Local-Area Network) connects to workstations, PC's, or other LANs to enable data access and device sharing.  Wake-on-LAN refers to ability to revive a system from sleep mode over a network without physically touching the system. [top]

Latency is the amount of time that one part of a system spends waiting for another part to catch up.  This is most common when the system sends data out to a peripheral device, and is waiting for the peripheral to send some data back (peripherals tend to be slower than onboard system components). [ top]

LBA (Logical Block Addressing) - A method used with SCSI and IDE drives to translate the cylinder, head and sector specifications of the drive to those usable by an enhanced BIOS.  LBA is used with drives that are larger than 528MB and causes the BIOS to translate the drives logical parameters to those usable by the system BIOS. [top]

MCH The Memory Controller Hub (MCH) provides the system bus interface, memory controller, AGP interface, and hub interface for I/O. [top]

MEC Commonly found in high-performance server and workstation designs where maximum memory capacity is required, a Memory Expansion Card (MEC) is used to allow for maximum memory slot availability (8, 12 or even 16 memory sockets), without sacrificing motherboard real estate. [ top]

Micro ATX The microATX form factor was developed as a natural evolution of the ATX form factor to address new market trends and PC technologies. While offering the same benefits of the ATX form factor specification, the microATX form factor improves upon the previous specification in several key areas. Current trends in the industry indicate that users require a lower-cost solution for their PC needs. This form factor addresses the cost requirement by reducing the size of the motherboard. The smaller motherboard is made possible by reducing the number of I/O slots supported on the board. The overall effect of these size changes reduces the costs associated with the entire system design. [ top]

Microprocessor - A processor on the RAID card which performs all RAID management functions (for example, the Intel i960).  Microprocessors offer higher performance compared to co-processors. [ top]

Mirroring (RAID 1) provides data protection by duplicating all data from a primary drive on a secondary drive. [top]

NOS (Network Operating System), i.e. NetWare, Windows NT server. [top]

NVRAM ROM and EEPROM are both examples of Non- Volatile RAM, memory that holds its data without power.  DRAM, in contrast, is volatile. [top]

OEMs (Original Equipment Manufacturers) refers to companies such as Compaq or IBM that package other companies' motherboards and hardware inside their case and sell them.

Online Capacity Expansion (O.C.E.) - A process for adding storage capacity to an existing RAID array without having to take the server offline.  Also known as Dynamic Array Expansion. [top]

OnNow , a term for a PC that is always on but appears off and that responds immediately to user or other requests. [top]

Overclocking - The process of running a processor, memory or video card at a speed faster than the officially marked speed by using a higher clock multiplier or faster bus speed.  Not recommended or endorsed by the device manufacturers.

The parallel port transmits the bits of a byte on eight different wires at the same time (that is, in parallel form, eight bits at the same time). [ top]

Parity - A method of error checking in which an extra bit is sent to the receiving device to indicate whether an even or odd number of binary 1 bits were transmitted.  The receiving unit compares the received information with this bit and can obtain a reasonable judgment about the validity of the character.  The same type of parity (even or odd) must be used by two communicating computers or both may omit parity.  When parity is used, a parity bit is added to each transmitted character.  The bit's value is 0 or 1, to make the total number of 1's in the character even or odd, depending on which type of parity that is used. [ top]

Partition - A section of a hard drive devoted to a particular OS or file system. [top]

PC99 is the 1999 - 2000 requirements for PC system and peripheral design for the "Designed for Microsoft Windows" logo.  Such as ACPI support and NO ISA slots. [top]

PCI (Peripheral Component I nterconnect) is a 32-bit local bus (data pathway) which is faster than ISA bus.  Local buses are those that operate within a single system (as opposed to a network bus, which connects multiple systems). [top]

The PCI PIO (PCI Programmable Input / Output) modes are the data transfer modes used by IDE drives.  These modes use the CPU for data transfer (DMA channels do not).  PCI refers to the type of bus used by these modes to communicate with the CPU. [top]

PCI-to-PCI bridge allows you to connect multiple PCI devices onto one PCI slot. [top]

The PCI-X specification addresses the need for increased bandwidth of PCI devices. PCI-X enables the design of systems and devices that operate at clock speeds up to 133MHz, or 1GB per second. Furthermore, PCI-X protocol enhancements enable devices to operate much more efficiently, thereby providing more useable bandwidth at any clock frequency.

Peer-to-Peer - A type of network in which any computer can act as both a server (by providing access to its resources to other computers) and a client (by accessing shared resources from other computers). [top]

Pentium - An Intel microprocessor with 32-bit registers, a 64-bit data bus and a 32-bit address bus.  The Pentium has a built in L1 cache that is segmented into a separate 8KB cache for code and another 8KB cache for data.  The Pentium includes a FPU or math co-processor.  The Pentium is backwards compatible with the x486 and can operate in real, protected virtual and virtual real modes. [ top]

Pentium Pro - An Intel 6th generation processor with 32-bit registers, a 64-bit data bus and a 36-bit address bus.  The Pentium Pro has the same segmented L1 cache as the Pentium but also includes a 256KB, 512KB or 1MB of L2 cache on a separate die inside the processor package.  The Pentium Pro is backwards compatible with the Pentium and can operate in real, protected and virtual real modes. [ top]

Pentium II - An Intel 6th generation processor similar to the Pentium Pro, but with MMX capabilities and SEC cartridge packaging technology. [ top]

Pipeline burst SRAM is a fast secondary cache.  It is used as a secondary cache because SRAM is slower than SDRAM, but usually larger.  Data is cached first to the faster primary cache, and then, when the primary cache is full, to the slower secondary cache. [ top]

Pipelining improves system performance by allowing the CPU to begin executing a second instruction before the first is completed.  A pipeline can be likened to an assembly line, with a given part of the pipeline repeatedly executing a set part of an operation on a series of instructions. [ top]

Plug and Play (PnP) - A hardware and software specification developed by Intel that allows a PnP system and PnP adapter cards to automatically configure themselves.  PnP cards are free from switches and jumpers and are configured via the PnP BIOS in the host system or via supplied programs for non-PnP systems. [top]

PM timers (Power Management timers ) are software timers that count down the number of seconds or minutes until the system times out and enters sleep (suspend) or doze mode.

PnP is an acronym for Plug and Play, a design standard that has become ascendant in the industry.  Plug and Play devices require little set-up to use.  Novice end users can simply plug them into a computer that is running on a Plug and Play-aware operating system (such as Windows 95), and go to work.  Devices and operating systems that are not Plug and Play require you to reconfigure your system each time you add or change any part of your hardware. [ top]

Power Management - Systems used initially in mobile computers (and now also used in desktop systems) to decrease power consumption by turning off or slowing down devices during periods of inactivity. [ top]

PXE (Preboot Execution Environment) is one of the four components that makes up the Wired for Management 2.0 baseline specification.  PXE was designed to define a standard set of preboot protocol services within a client, towards the goal of allowing networked - based booting to boot using industry standard protocols. [top]

RAID (Redundant Array of Independent Disks) is a way of storing same data in different places on many hard drives.  By using this method, the data is stored redundantly, also the multiple hard drives will appear as a single drive to the operating system.  RAID level 0 is known as striping, where data is striped (or overlapped) across multiple hard drives, but offers no fault-tolerance.  RAID level 1 is known as mirroring, which stores the data within at least two hard drives, but does not stripe.  RAID level 1 also allows for faster access time and fault-tolerance, since either hard drive can be read at the same time.  RAID level 0+1 is both striping and mirroring, providing fault-tolerance, striping, and faster access all at the same time. [ more on RAID ]

RAID Levels (0 through 5) refer to different array architectures that offer various advantages in terms of data availability, cost and performance.  RAID levels 0, 1, 0/1, and 5 are the most popular. [top]

RAID 0 - See "Striping".

RAID 0/1 - Combines RAID 0 (data striping) and RAID 1 (disk mirroring). [top]

RAID 1 - See "Mirroring".

RAID 5 - Combines data striping (for enhanced performance) with distributed parity (for data protection) to provide a recovery path in case of failure. [top]

RAID Management Software makes installation, configuration, and management of RAID arrays easy.  Often includes features such as pager notification and remote management.

The term RAM (Random Access Memory), while technically referring to a type of memory where any byte can be accessed without touching the adjacent data, is often used to refer to the system's main memory.  This memory is available to any program running on the computer. [top]

RDRAM Developed by Rambus, Inc., RDRAM, or Rambus DRAM, is currently the fastest memory technology used by PCs. While today's most common SDRAM delivers data at a maximum speed of 133MHz, RDRAM currently transfers data at up to 800MHz. RDRAM is expected to be seen in 1066MHz and 1200MHz iterations in the coming months and years.

Registry - The system configuration files used by Windows 9x and Windows NT to store settings about installed hardware and drivers, user preferences, installed software and other settings required to keep Windows running correctly.  Replaces the Win.ini and System.ini files from the older Windows 3.x [ top]

ROM (Read-Only Memory) is a storage chip which contains the BIOS (Basic Input / Output System), the basic instructions required to boot the computer and start up the operating system. [top]

SCSI (Small Computer System Interface) is the technology that allows you to connect various devices to your PC.  This connection is made using a SCSI card that fits inside your computer. [ top]

SCSI-1 is a complete document with all the physical and protocol Layers, it is obsolete. 8 bit SCSI, Single-ended open collector drivers with Asynchronous transfer for all commands and data transfers. Bipolar technology for the drivers and receivers, bus DC loads a major consideration, passive termination. The connectors are the low density 50 pin internal and external connectors, the external low density connector is also known as the Centronics connector, in Cable terms the SCSI-1 Connector. [ top]

SCSI-2 is a complete document with all the Physical and protocol layers. SCSI asynchronous commands with Synchronous data transfer rates up to Fast 10 megatransfers per second or 20 megabytes per second. Bipolar technology with Open collector or active negation single ended bus transceivers, passive or active termination, and Differential (Now called HVD High Voltage differential) bus transceivers based on EIA 485 allowed in SCSI-2. The A connectors are the 50 pin low density internal connector, low density Centronics external connector and the High density external connector. Cable terms the SCSI-2 connector is the high density 50 pin. The B connectors for wide bus was not practical because it required a second cable 68 pin High density. [ top]

SCSI-3 change the document structure, SCSI-3 is not one document with all the different layers and electrical interfaces, but a collection of documents that cover the physical layer, the basic protocol specific to that electrical interface, the primary command set layer (SPC) and the specific protocol layer. The specific protocol layer contains the Hard Disk interface Commands in the Block Commands (SBC), Stream Commands for tape drives(SSC), Controller Commands for RAID arrays (SCC), Multimedia Commands (MMC), Media Changer Commands (MCC) and enclosure services commands (SES) for example. There is an overall architectural model (SAM). Each document has its own revision level, these are normally referred to as SCSI, the 3 has been dropped. [ top]

SCSI Parallel interface (SPI) defines the Parallel bus electrical connections and Signals.

SCSI Interlock Protocol (SIP) defines the parallel command set. SIP is include in the SPI-2 and SPI-3 revisions. SPI define the P connector 68 pin High density primary cable for a wide bus in a single cable. This 68 pin High density cable is referred to as the SCSI-3 connector. SPI includes speeds up to Fast SCSI (Wide Fast SCSI is 20 MegaBytes per second), electrical interfaces are defined in CMOS terms, with the termination specified at 0.2 volts instead of 0.5 volts. The A cable is referenced to SCSI-2 but not included in SPI. [ top]

SDRAM (Synchronous Dynamic RAM) is so-called because it can keep two sets of memory addresses open simultaneously.  By transferring data alternately from one set of addresses, and then the other, SDRAM cuts down on the delays associated with non-synchronous RAM, which must close one address bank before opening the next. [top]

The Serial Port is so called because it transmits the eight bits of a byte of data along one wire, and receives data on another single wire (that is, the data is transmitted in serial form, one bit after another). [top]

SIMM (Single In-line Memory M odules) are the most common form of RAM.  They must be installed in pairs, and do not have the carrying capacity or the speed of DIMMs. [top]

SIMM bank/socket SIMM sockets are the physical slots into which you stick SIMM modules.  A pair of SIMM sockets forms a SIMM bank, and act as a unit.  If only one socket is filled, the bank will not operate. [top]

In Sleep / Suspend mode, all devices except the CPU shut down. [top]

SMA (Shared Memory Architecture) with system memory. [ top]

SMbus (System Management Bus) is a two-wire interface based on the I2C protocol. It is a low-speed bus that provides positive addressing for devices as well as bus arbitration. [top]

SMP (Symmetric MultiProcessing) is a computer architecture that provides fast performance by making multiple processors available to complete individual processes simultaneously. Unlike asymmetrical processing, any idle processor can be assigned any task, and additional processors can be added to improve performance and handle increased loads. A variety of today's operating systems and hardware configurations are available to support SMP. Specific applications can benefit from SMP if the code provides for multithreading. SMP uses a single operating system and shares common memory and disk input/output resources. Windows NT, Windows 2000, Windows XP Professional and many flavors of Linux and UNIX support SMP. [top]

SPI-2 doubles the speed again to Fast-40 commonly know as Ultra2 SCSI, 40 megatransfers per second or on a wide bus 80 megabytes per second with a new electrical interface, low voltage differential, LVD SCSI. Single ended can not be used for speeds above Fast-20. The SPI-2 standard included SIP, the Single Connector Attachment (SCA-2) 80 pin Host swap connector and the 68 pin Very High Density Connector (VHDCI). The SCSI-2 A and SPI P connectors are include in SPI-2. [ top]

SPI-3 doubles the speed again to Fast-80DT commonly know as Ultra160 SCSI, 80 megatransfers per second or on a wide bus 160 megabytes per second. There are a lot of changes with SPI-3; High voltage differential (HVD SCSI) and the 32 bit data bus with the Q cable are obsolete, clocking is defined for LVD on both the rising and falling edges of the REQ/ACK clock Double Transition (DT) defined for wide bus only, Cyclic Redundancy Check (CRC) defined, Domain Validation defined part of the domain validation will be in SPC-x, Packetized Commands and Messaging defined and Quick Arbitration Defined. [ top]

SRAM Static RAM, unlike DRAM, does not need to be refreshed in order to prevent data loss.  Thus, it is faster, and more expensive. [top]

SSI Intel is currently working with leading server OEMs and industry suppliers to help gain ground for the SSI form factor initiative. The goal of the SSI initiative is to deliver a set of specifications covering two primary server elements: power supplies and electronic bays, and thus lay the groundwork for future growth in the server market. The initiative is similar to such desktop PC efforts as ATX and NLX, in that it is working to define common packaging elements for the server market. [ top]

In Standby mode, the video and fixed disk drive shut down; all other devices operate normally. [top]

Striping (RAID 0) - Spreads data evenly over multiple drives to enhance performance.  Because there is no redundancy scheme, it does not provide data protection. [top ]

UltraDMA - A protocol for transferring data to an ATA interface Hard Drive.  The UltraDMA/33 protocol transfers data in burst mode at a rate of 33Mbytes/sec, while the even faster UltraDMA/66 protocol transfers at 66Mbytes/sec.  UltraDMA/66 also requires the use of a special 80-pin conductor cable for signal integrity. This cable is also recommended for UltraDMA/33 and is backwards compatible with standard ATA IDE cables.

USB (Universal Serial Bus) is a versatile port.  It can function as a serial, parallel, mouse, keyboard, or joystick port.  It is fast enough to support video transfer, and is capable of supporting up to 127 daisy-chained peripheral devices. [top]

VCM (Virtual Channel Memory) is the new SDRAM architecture, which realizes flexible and high-efficiency data transfer by incorporating channel buffers configured by high-speed registers. [top]

VGA (V ideo Graphics Array) is the PC video display standard. [top]

Virtual Memory - A technique by which operating systems load more programs and data into memory than they can hold.  Parts of the programs and data are kept on disk and constantly swapped back and forth into system memory.  The applications' software programs are unaware of this setup and act as though a large amount of memory is available.

VRM (Voltage Regulator Module) regulates the voltage fed to the microprocessor. [top]

V-SYNC controls the vertical properties of the monitor. [top]

VxD (Virtual Device Driver) - A special type of Windows driver.  VxDs run at the most privileged CPU mode (ring 0) and allow low-level interaction with the hardware and internal Windows functions. [top]

WfM (Wired for Management) is an Intel guideline on the implementation of system management. [top]

WHQL (Windows Hardware Quality Libs) is the test procedure for "Designed for Microsoft Windows" logo. [top]

ZIF sockets (Z ero Insertion Force sockets) make it possible to insert CPUs without damaging the sensitive pins.  The CPU is lightly placed in an open ZIF socket, and the metal lever pulled down.  This shifts the processor over and down, guiding it into place on the board. [top]

 Information published on is subject to change without notice.
Copyright © 2006, Tyan Computer Corporation. All Rights Reserved.