Glossary

10 Gigabit Ethernet — An Ethernet standard, based on the IEEE 802.11ae and 802.11ak standards, for 10 Gbps transmission. See also Ethernet and IEEE 802 standards.

24-bit color — A color display scheme that represents each pixel�s color as three 8-bit numbers, each representing the intensity of an additive or a subtractive color; results in chromatic depth of around 16 million colors. See also chromatic depth.

802.11 — An early IEEE wireless networking standard that defines a 2 Mbps maximum raw data transfer rate in the 2.4 GHz band.

802.11a — An IEEE wireless networking standard that divides frequency bands in the 5.2, 5.7, and 5.8 GHz ranges into 12 channels; uses orthogonal frequency division multiplexing (OFDM) to achieve standard transmission speeds of 6, 9, 12, 18, 24, 36, 48, and 54 Mbps.

802.11b — An IEEE wireless networking standard that divides the 2.4 GHz band into 14 channels, each with 22 MHz of bandwidth; uses direct sequence spread spectrum (DSSS) to yield raw data transfer rates of 22, 44, and 88 Mbps.

802.11g — An IEEE wireless networking standard that combines the frequencies and bit-encoding methods of 802.11b with the OFDM transmission method of 802.11a; raw transmission speeds are the same as 802.11a.

802.11n — An IEEE wireless networking standard that expands on 802.11g; capable of broadcasting or receiving on up to four frequencies in the 2.4 or 5 GHz bands.

80x86 processors — A family of processors that enhanced the capabilities of the original 8088/8086 processors and enabled Microsoft OSs to develop beyond MS-DOS.

absolute addressing — In programming, using memory address operands that refer to actual physical memory locations; this method requires knowing process offsets. See also process offset.

access arm — In magnetic disks, the device where read/write heads are mounted; it�s attached to a positioning servo for placing read/write heads on specific tracks.

access control list (ACL) — A list describing rights granted or denied to users, groups, and computers for accessing network resources.

access point (AP) — A device that connects a wireless network to a wired network and manages media access, performs error detection, and implements security protocols for the wireless network.

access time — The time required to perform one complete read or write operation; a measure of storage device time.

acknowledge (ACK) — An ASCII control character sent by a receiver if no data errors are detected.

Active Directory — A directory service and security system built into Windows Server. See also directory services.

active matrix display — An LCD that uses one or more transistors for every pixel.

ADD — An instruction that accepts two numeric inputs and produces their arithmetic sum.

additive colors — The primary colors for video display (red, green, and blue).

address — The location of a data element in a storage device; often used in data structures.

address bus — The portion of the system bus that transmits a memory address when primary storage is the sending or receiving device. See also system bus.

address mapping — The process of the CPU determining the physical memory address that corresponds to a memory reference.

address resolutionSee address mapping.

addressable memory — The highest numbered storage byte that can be represented in a CPU or computer; usually determined by the number of bits used to represent an address.

Advanced Intelligent Tape (AIT) — A magnetic tape standard developed by Sony based on Digital Audio Tape; uses helical scanning and an improved tape drive technology to pack more data onto a single tape. AIT includes a small RAM cache in cartridges, which stores directory information to speed searching and data access. See also Digital Audio Tape (DAT) and helical scanning.

algorithm — A program in which different instructions are applied to different data input values, depending on the outcome of decisions the program makes. This term also applies to processing steps that describe the solution to a problem.

allocation unit — The smallest number of secondary storage bytes that can be allocated to a file; can�t be smaller than the unit of data transfer between the storage device and device controller.

American Standard Code for Information Interchange (ASCII) - A standard 7-bit coding method for character data and some device control codes.

amplifier — A device that increases a signal�s amplitude and can extend a signal�s range by boosting signal power to overcome attenuation; a drawback is that any existing noise or distortion in the signal is amplified as well. See also attenuation and repeater.

amplitude — A measure of wave height or power; the maximum distance between a wave�s peak and its zero value.

amplitude modulation (AM) — A modulating method that represents bit values as specific wave amplitudes.

amplitude-shift keying (ASK) See amplitude modulation.

analog signal — A signal that uses the full range of a carrier wave characteristic to encode continuous data values; because it�s continuous in nature, it can represent any data value within a range of values.

analog-to-digital converter (ADC) — An audio device that accepts a continuous electrical signal representing sound, samples it at regular intervals, and outputs a bitstream representing the samples. See also sampling.

AND — An instruction that generates the result true only if both its data inputs are true.

applet — A Java program that runs inside another program, such as a Web browser, and performs functions such as accepting user input and displaying forms and images. See also Java.

application development software — Programs used to develop other programs, including application software, system software, and other application development programs; encompasses compilers and interpreters for programming languages and integrated software development packages.

application firewall — A server that handles service requests from external users of applications; improves security by shielding internal servers and resources from direct access by outside users.

Application layer — The OSI layer that includes communication protocols used by programs that make and respond to high-level requests for network services. See also Open Systems Interconnection (OSI) model.

application software — A stored set of instructions for responding to a specific request or performing a specific task.

architectural design — The first set of activities in the UP�s design discipline; involves selecting and describing the configuration of all hardware, network, systems software, and application development tools to support system development and operations. See also design discipline and Unified Process (UP).

areal density — The surface area allocated to a bit on a storage medium, typically measured in bits, bytes, or tracks per inch; also called �recording density� or �bit density.�

arithmetic logic unit (ALU) — One of the main CPU components, containing circuitry for performing computation, comparison, and logic instructions. See also central processing unit (CPU).

arithmetic SHIFT — A SHIFT instruction that performs division or multiplication. See also SHIFT.

array — An ordered list of data elements, in which each element can be referenced by an index to its position; array elements are normally referenced by the array name and the index value.

assembler — A program that translates an assembly-language program into binary CPU instructions; the earliest example of automated program development tools. See also assembly language.

assembly language — A programming language that uses mnemonics to represent CPU instructions and memory addresses; also called a second-generation language (2GL).

Association for Computing Machinery (ACM) — A professional society for computer scientists, programmers, and engineers.

Association for Information Technology Professionals (AITP) — A professional society for information system managers and application developers.

asynchronous transmission — A method in which messages are sent on an as-needed basis, so sender and receiver don�t synchronize their clocks continuously.

attenuation — A reduction in signal amplitude caused by interactions between the signal�s energy and the transmission medium; proportional to the medium�s length.

audio response unit — A device that generates spoken messages based on text input; commonly used for automated phone bank tellers and automated call routing.

auditing — The process of creating and managing records of user activity or resource access.

authentication — The process of determining or verifying the identity of a user or process owner.

authorization — The process of determining whether an authenticated user or process has enough rights to access a resource.

average access time — Typically expressed as an average of access times for all storage locations. See also access time.

back-end CASE tool — An application development tool that generates source code instructions based on analysis and design models; also called a code generator. See also computer-assisted software engineering (CASE) tool.

bandwidth — The difference between maximum and minimum frequencies that be transmitted through a transmission medium.

bar code — A series of vertical bars of equal length but varied thickness and spacing, used to encode numeric data.

bar-code scanner — An optical input device that detects specific patterns of bars or boxes representing numeric data.

bare-metal hypervisor — A hypervisor that is installed much like an OS but provides only minimal OS-type functions, such as the capability to start and stop virtual machines (in contrast to a hypervisor installed above a traditional operating system). See also hypervisor and virtualization environment.

base — A multiplier that describes the difference between one position and the next in a numbering system.

benchmark — A measure of CPU or computer performance when carrying out one or more specific tasks; used to compare the performance of multiple computers, measure a computer�s performance and determine how to improve it, and determine the computer or system configuration that meets an application�s requirements.

benchmark program — A program that performs specific tasks that can be counted or measured. See also benchmark.

benchmark suite — A collection of benchmark programs for evaluating computer systems. See also benchmark.

big endian — A CPU or memory architecture in which the most significant byte is stored at the lowest memory address. See also most significant byte.

billions of floating-point operations per second (gigaflops or GFLOPS) — A measurement of the rate at which floating-point operations are performed; used to measure CPU performance.

binary number — A number in which each digit can have only one of two possible values (0 or 1).

binary signals — Digital signals in which one of two values is encoded by modulating a wave characteristic. See also digital signal.

biometric authentication — A method of verifying identity based on physical characteristics, such as fingerprints and facial features. See also authentication.

bit — Derived from the term �binary digit,� it represents one digit of a binary number and can have the value 0 or 1.

bitmap — A stored set of numbers describing the content of all pixels in an image.

bit string — A group of bits describing a single data value.

bit time The duration of each bit in a carrier signal.

bladeA circuit board containing most of a server but lacks secondary storage, external I/O connections, and a power supply; concentrates more computing power in less space and with lower power requirements than a typical cluster.

block — A generic term for describing secondary storage data transfer units. Also refers to a set of logical records grouped on a storage device for efficient processing, storage, or transport as well as a portion of a program that�s always executed as a unit.

block check character (BCC) — The combined parity bits from each position in a group of characters or bytes; added to the end of the block before transmitting. See also parity bit.

block checking — An error-checking method for groups of characters or bytes in which the sender combines parity bits for each position into a block check character (BCC) and adds it to the end of the block. See also parity bit.

blocked state — The state of an active thread that�s been suspended by the OS and is waiting on the stack until interrupt processing has been completed. See also interrupt and thread.

blocking — The grouping of logical records in physical records. See also logical record and physical record.

blocking factor — A numeric ratio of logical records to physical records. See also logical record and physical record.

Blu-ray disc (BD) — An update to DVD-ROM, originally designed for high-definition video discs but has been adapted to data storage. See also DVD read-only memory (DVD-ROM).

Boolean data type — A data type that can store only the value true or false; requires only a single bit for storage.

Boolean logic — A formal logic system in which statements can be evaluated only as true or false; well suited to the binary numbers used in computer processing.

BRANCH — An instruction that causes the processor to depart from sequential instruction order; its operand is loaded into the register that the control unit uses to fetch the next instruction.

branch prediction — An approach to dealing with conditional BRANCHes in which the CPU guesses whether a branch condition will be true or false based on past experience; a form of parallel processing.

broadband — A high-bandwidth communication channel.

broadcast mode — A communication mode in which the same message is transmitted to all devices on a network simultaneously. See also simplex mode.

brownout — A temporary reduction in voltage level by a power provider, usually because of demand for electricity exceeding the provider�s generation and transmission capabilities during peak demand periods.

buffer — A small reserved area of main memory (usually DRAM or SRAM) that holds data in transit from one device to another and is used to resolve differences in data transfer unit size.

buffer overflow — An error condition that results when receiving more data than can be stored in a buffer. See also buffer.

bus — A shared electrical or optical communication channel that connects two or more devices in a computer or network.

bus arbitration unit — A simple processor attached to a peer-to-peer bus that decides which devices must wait when multiple devices want to become a bus master. See also bus master and peer-to-peer bus.

bus clock — A clock circuit that generates timing pulses, which are transmitted to all devices attached to the system bus to coordinate their activities.

bus cycle — The time interval from one bus clock pulse to the next; also the time required to perform one data transfer operation on a bus.

bus master — A device attached to a bus that can initiate a data transfer operation or send a command to another device; it controls all access to the bus.

bus protocol — A communication protocol, used by all devices attached to a bus, that governs the format, content, and timing of data, memory addresses, and control messages sent across the bus.

bus slaves — Devices that must go through the bus master for access to the bus. See also bus master.

bus topology — A network topology in which every node is directly connected to a single shared transmission line. See also bus and network topology.

business logic layer — The software layer that carries out the rules and procedures of business processing. See also three-layer architecture.

business modeling discipline — Activities in the Unified Process for developing models of an organization and the system environment. See also Unified Process (UP).

business-to-business (B2B) — Interconnection of a company�s information systems with customer and supplier information systems to improve efficiency.

byte — A string of 8 bits; generally the smallest unit of data that can be read from or written to a storage device. See also bit.

cache — An area of high-speed memory (usually RAM) for storage device accesses that improves the performance of read and write operations.

cache controller — A special-purpose processor or software that manages cache content; it guesses what data will be requested in the near future and loads this data from the storage device into the cache before it�s actually requested.

cache hit — An access to data already contained in the cache.

cache miss — An access to data that isn�t stored in the cache.

cache swap — An operation performed after a cache miss. The cache controller guesses which data is least likely to be needed in the near future, writes it to the storage device, and purges it from the cache. The requested data is then read from the storage device and placed in the cache. See also cache miss.

call instruction — In programming, an instruction that transfers control to the first instruction in a function. See also function.

capital expenditures — Funds an organization uses on capital resources. See also capital resources.

capital resources — Assets or resources expected to provide benefits beyond the current fiscal year.

Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA) — A MAC protocol, used in wireless networks, that uses a three-step carrier sense and transmission sequence to try to avoid collisions. See also collision and Media Access Control (MAC).

Carrier Sense Multiple Access/Collision Detection (CSMA/CD) — A MAC protocol that allows collisions to occur but has methods for detecting and recovering from them. See also collision and Media Access Control (MAC).

carrier wave — A wave with encoded bits in a communication channel.

Category 5 — Similar to Category 6 but can�t achieve the same transmission speeds reliably.

Category 6 — The most widely used twisted-pair wiring standard; consists of four twisted pairs that transmit at speeds up to 1 Gbps.

cathode ray tube (CRT) — An older video display device that�s an enclosed glass vacuum tube with an electron gun generating a narrow beam of electrons toward the tube�s front surface, which is coated with colored phosphors that emit light when struck by electrons.

CD digital audio (CD-DA) — A read-only format for storing and distributing music on a CD.

CD read-only memory (CD-ROM) — A standard 120-mm read-only optical disc; compatible with CD-DA but includes additional formatting information to store directory and file information. See also CD digital audio (CD-DA).

central processing unit (CPU) — A general-purpose processor that executes all instructions and controls all data movement in a computer system. See also general-purpose processor.

channelSee I/O channel or communication channel.

character — A symbol in a written language, including letters, numerals, and punctuation marks.

character-framing methods — Approaches to clock synchronization when messages consist of ASCII or Unicode characters.

chief information officer (CIO) — The person who�s responsible for planning, maintaining, and operating all information-processing resources in an organization; managers such as the database administrator, network administrator, and computer operations manager often report to the CIO.

child — The file version that�s been updated with changes to the parent file. See also parent.

child process — A process created and controlled by the parent process that spawned it. See also process and spawn.

chromatic depth — The number of distinct colors or gray shades that can be displayed in a grayscale image. See also grayscale.

chromatic resolutionSee chromatic depth.

circuit switching — A channel-sharing strategy that grants exclusive use of a communication channel for the duration of the session.

class — A data structure containing both traditional (static) data elements and programs that manipulate data; it combines related data items in much the same way a record does, but it extends a record to include methods for manipulating data items.

clear-to-send (CTS) signal — A CSMA/CA signal that a wireless access point transmits after it detects no potential collision. See also Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA).

client — A program or computer that requests services from another program or computer.

client/server architecture — A method of organizing software to provide and access distributed information and computing resources; divides software into two classes�client and server.

clock cycle — The time interval between two clock timing pulses. See also system clock.

clock rate — The frequency (expressed in Hz) at which the system clock generates timing pulses. See also system clock.

cloud — A specific way of organizing computing resources for maximum availability and accessibility and minimum complexity in the interface; includes front-end Web-based interfaces and a large collection of computing and data resources (collectively called �back-end resources�).

cloud computing — A cloud-based approach to distributing and accessing software and hardware services across the Internet.  See also cloud, infrastructure as a service, platform as a service, and software as a service

cluster — A group of similar or identical computers, connected by a high-speed network, that cooperate to provide services or run a common application; has the advantages of scalability and fault tolerance but can be complex to configure and manage.

CMY — Cyan, magenta, and yellow; the primary colors in printing. See also subtractive colors.

CMYKCyan, magenta, yellow, and black; represents the primary colors in printing plus a separate ink for black.

coaxial cable — A transmission medium that contains a single copper conductor surrounded by a thick plastic insulator, a metallic shield, and a tough plastic outer wrapping.

code — Program instructions for performing a task.

coercivity — The capability of substance or magnetic storage medium to accept and hold a magnetic charge; directly proportional to mass.

collating sequence — The specific order for assigning numeric codes to charactersof symbols.

collision — The noise or interference produced when multiple nodes attempt to transmit across the same medium at the same time, and their messages mix.

colon hexadecimal notation — The written format of 128-bit IPv6 addresses; written in the form hhhh:hhhh:hhhh:hhhh:hhhh:hhhh:hhhh:hhhh, with hhhh representing a sequence of four hexadecimal digits. See also Internet Protocol version 6 (IPv6).

command language — A set of commands and syntax requirements for implementing an operating system�s command layer via a text interface.

command layer — The operating system layer that serves as the user interface; via this layer, users can run applications and OS utilities and manage system resources, such as files, folders, and I/O devices.

Common Object Request Broker Architecture (CORBA) — An industry-wide interoperability standard specifying the middleware that objects use to interact across networks.

communication channel — A combination of a sending device, a receiving device, the transmission medium connecting them, and a communication protocol. See also communication protocol and transmission medium.

communication protocol — A set of rules and conventions for representing the content of data and commands, encoding and transmitting data (bits), channel organization, and communication coordination (clock synchronization and error detection and correction).

compact disc (CD) — A technology developed by Sony and Phillips for storing and distributing music in the CD-DA format on a 120-mm optical disc. See also CD digital audio (CD-DA).

compaction — The process of reallocating all programs in memory so that free partitions form a contiguous block in upper memory; used to address the problem of fragmentation. See also contiguous and fragmentation.

competitive advantage — The way in which an organization uses resources to offer better or cheaper services so that it has a major edge over its competitors.

compiler — A program that translates some source code instructions into executable code and others into library calls; a compiler�s output is called object code. See also object code.

compiler library — A file containing related executable functions and an index of the library contents. See also compiler and link editor.

complete path — The access path that begins at the root directory and proceeds through all directories along a path to the file being accessed.

complex instructionsInstructions that combine primitive processing operations.

complex instruction set computing (CISC) — A computer and processor design approach, using complex instructions that do more work per instruction; it reduces the extra memory required for program storage and execution in RISC CPUs. See also reduced instruction set computing (RISC).

component — A standardized, interchangeable software module that�s executable, has a unique identifier, and has a well-known interface.

Component Object Model Plus (COM+) — A Microsoft specification for component interoperability; defines component registration, message-routing services, and a component communication protocol.

composite signal — A complex signal created by combining multiple simple signals.

compression — A technique that reduces the number of bits used to encode data.

compression algorithm — A mathematical compression technique implemented as a program for translating data inputs into equivalent but smaller data outputs. See also compression.

compression ratio — The ratio of data size in bits or bytes before and after compression. See also compression.

computer-assisted software engineering (CASE) tool — An application development tool suite that supports the Unified Process requirements and design disciplines. See also design discipline and requirements discipline.

computer network — A collection of hardware and software components that enable users and computer systems to share information, software, and hardware resources and make it possible to use many types of communication methods.

computer operations manager — The person who oversees the operation and maintenance of a large information-processing facility or an information system.

computer science — The study of implementing, organizing, and applying computer software and hardware resources.

concurrent execution — A method of sharing CPU control among threads by using time slices. See also thread.

condition A comparison or other logical operation that produces a Boolean (true or false) result.

conditional BRANCH — A BRANCH instruction that occurs only if a specified condition is met. The condition is evaluated, and the Boolean result is stored in a register; the register�s contents are checked, and the BRANCH is performed only if the Boolean result is true.

conductivity — The capability of an element or a substance to enable electron flow.

conductor — A substance that exhibits conductivity, allowing electrons to flow through it.

connectionless protocol — A communication protocol that doesn�t require sender and receiver to establish a connection before transmitting any data.

connection-oriented protocol — A communication protocol that requires sender and receiver to establish a connection before transmitting any data.

contiguous — The condition of all portions of a program or the OS being loaded into sequential physical locations in memory. See also memory allocation.

control bus — The portion of the system bus that carries commands, command responses, status codes, and similar messages; computer components coordinate their activities by sending signals over this bus. See also system bus.

control structure — A source code instruction that controls the execution of other source code instructions; includes unconditional BRANCHes, such as a goto statement; conditional BRANCHes, such as an if-then-else statement; and loops, such as while-do. See also source code.

control unit — One of the main CPU components, responsible for moving data, accessing instructions, and controlling the arithmetic logic unit. See also central processing unit (CPU).

core A term describing the logic, computation, and control circuitry of a single CPU. See also multicore architecture.

core memory In early computers, a technology for implementing primary storage as rings of ferrous materials embedded in a two-dimensional wire mesh.

crosstalk — In parallel transmission channels, noise added to the signal in a wire from EMI generated by adjacent wires. See also parallel transmission.

current directory — The directory that�s being accessed. See also directory.

cursor — A symbol on a video display that indicates the current position; also called a �pointer.�

cycle — In communication, the full range of a sine wave, from zero to positive peak, back to zero, to negative peak, and back to zero again.  In CPU operation, the sequence of low-level actions performed to fetch and instruction from memory and execute it.

cycle time — The inverse of the clock rate; in most CPUs, it�s the time required to fetch and execute the simplest instruction in the instruction set. See also clock rate.

cyclic redundancy checking (CRC) The most widely used error-detection method; uses a complex algorithm to generate CRC bit strings for groups of characters or bytes.

cylinder In magnetic disks, consists of all tracks at an equivalent distance from the edge or spindle on all platter surfaces. See also platters and track.

data bus — The portion of the system bus that transmits data between computer components. See also system bus.

data declaration — A source code instruction type that defines the name and data type of program variables. See also source code.

data layer — The software layer that manages stored data, usually in databases. See also three-layer architecture.

Data Link layer — The OSI layer serving as the interface between network software and hardware; responsible for media access control and converting messages and addresses from one format to another. See also Open Systems Interconnection (OSI) model.

data operation — A source code instruction type that updates or computes a data value, such as an assignment statement or a computation. See also source code.

data striping — A fault-tolerance technique that breaks a unit of data into smaller segments and stores these segments on multiple disks. See also fault tolerance.

data structure — A related group of primitive data elements organized for some type of common processing; it�s defined and manipulated in software because the CPU can�t manipulate data structures directly.

data transfer rate — The rate at which data is transmitted through a medium or communication channel; measured in data units per time interval; essentially, it�s a measure of communication capacity. For a storage device, it�s computed by dividing 1 by the access time and multiplying the result by the unit of data transfer.

database administrator — The person responsible for overseeing an organization�s database and ensuring data integrity, reliability, security, and availability.

datagrams — Messages accepted from Transport-layer protocols and forwarded to their destinations. See also Transport layer.

debugging tools — Utilities that simulate program execution and help programmers trace errors. See also application development software.

debugging version — A program version containing symbol table entries and debugging checkpoints to help locate and correct errors. See also production version and symbol table.

decimal point — The period or comma in the decimal numbering system that separates the whole and fractional parts of a numeric value. See also radix point.

decoding — The process the control unit performs of extracting an instruction�s op code and operands, loading data inputs, and signaling the ALU.

decompression algorithm — An algorithm that restores compressed data to its original or nearly original state. See also compression.

deployment discipline — Activities in the Unified Process for installing and configuring infrastructure and application software components and bringing them into operation. See also Unified Process (UP).

design discipline — Activities in the Unified Process for determining the structure of a specific information system that fulfills the system requirements. See also Unified Process (UP).

design models — Models that specify detailed blueprints for software component construction and the interaction between software components and users.

detailed design — Design activities that specify system details, including databases, application software, user and system interfaces, and backup and recovery mechanisms.

device controller — A processor that controls the physical actions of storage and I/O devices; connects these devices to the system bus or a subsidiary bus.

differential backup — A type of backup that�s a variation on an incremental backup, in which backup times aren�t reset as files are copied. See also incremental backup.

Digital Audio Tape (DAT) — An early magnetic tape technology on which Digital Data Storage standards are based. See also Digital Data Storage (DDS).

Digital Data Storage (DDS) Magnetic tape standards developed by Hewlett-Packard and Sony and based on Digital Audio Tape; DDS drives use helical scanning. See also Digital Audio Tape (DAT) and helical scanning.

digital signal — A signal that can contain one of a finite (countable) number of possible values.

digital signal processor (DSP) — A microprocessor specialized for processing continuous streams of audio or graphical data; commonly embedded in audio and video hardware.

digital-to-analog converter (DAC) — An audio device that accepts a bitstream representing sound samples and generates a continuous analog signal that can be amplified and routed to a speaker.

digitizer — A device consisting of a digitizing tablet and a pen, stylus, or both that captures a pointing device�s position as input data.

Direct3D — A video controller image description language that�s part of the Microsoft DirectX suite embedded in Windows OSs.

direct accessSee random access.

direct-attached storage (DAS) — A storage access model in which software running on a CPU accesses secondary storage devices in the same computer.

direct memory access (DMA) — A method of data transfer that enables the CPU to execute instructions while another device (the DMA controller) manages all transfers between memory and other storage or I/O devices. See also DMA controller.

directory — A data structure containing information about files and other directories.

directory services — Middleware that stores the name and network address of distributed resources, responds to directory queries, accepts directory updates, and synchronizes directory copies. See also middleware.

disciplines — Related groups of system development activities in the Unified Process. See also Unified Process (UP).

discrete signalSee digital signal.

disk defragmentation — Reorganizing data on a disk drive so that a file�s contents are stored in sequential sectors, tracks, and platters; an OS utility is used to perform this task.  See also fragmentation.

disk mirroring — A fault-tolerance technique in which all disk write operations are made simultaneously or concurrently to two storage devices. See also fault tolerance.

dispatching — Giving CPU control to a thread in the ready state. See also ready state and thread.

distortion — Changes to the data signal caused by interaction with the communication channel; can include echoes, resonance, and selective attenuation.

distributed computingSee distributed processing.

Distributed Computing Environment (DCE) — A standard for distributed OS services defined by the Open Group Standard; covers network directory services, file-sharing services, RPC, remote thread execution, system security, and distributed resource management. See also distributed processing and Remote Procedure Call (RPC).

distributed processing — Spreading parts of an information system across many computer systems and locations.

dithering — A process that generates continuous color approximations by placing small dots of different colors in an interlocking pattern.

diversity — As specified in the 802.11n standard, using antenna pairs for redundant data transmission across different frequencies to increase signal reliability. See also 802.11n.

DMA controller — A device attached to the system bus and main memory that manages data transfers, freeing the CPU to execute instructions. See also direct memory access (DMA).

Domain Name System (DNS) — A name-resolution protocol used on the Internet; makes use of dynamic connections to find requested IP addresses. See also dynamic connection.

dot matrix printer — An impact printer that moves a print head containing a matrix of pins over the paper, and a pattern of pins matching the character or symbol to be printed is forced out of the print head.

dots per inch (dpi) — A measure of print or display resolution (pixel density); a smaller pixel size represents a higher dpi and, therefore, higher image or print quality. See also resolution.

dotted decimal notation — The written format of IPv4 addresses; written in the form ddd.ddd.ddd.ddd, with ddd representing a decimal number between 0 and 255. See also Internet Protocol version 4 (IPv4).

double data rate (DDR) — A series of technologies, each doubling the data transfer rate of the previous synchronous DRAM version. See also synchronous DRAM (SDRAM).

double inline memory module (DIMM) — A small printed circuit board that�s essentially a SIMM with independent electrical contacts on both sides of the module. See also single inline memory module (SIMM).

double-precision — A data format that combines two adjacent fixed-length data items to hold a single value; increases accuracy or numeric range.

doubly-linked list — A data structure in which each list element contains pointers to both the previous and next list elements.

drive array — An arrangement of hard drives enclosed in a storage cabinet and access as though they�re a single storage device.

dual inline packages (DIPs) — An early form of packaging for RAM or ROM circuits; had two rows of electrical contact pins.

dual-porting — The simultaneous read/write capability in video RAM.

DVD — An optical disc format for distributing movies and other audiovisual content; stands for for both �digital video disc� and �digital versatile disc.�

DVD read-only memory (DVD-ROM) — A format for general-purpose read-only data storage on DVD.

dynamic connection — A more complex, but flexible, approach to remote resource access, in which connections between a client and a server or remote resource aren�t established until the time of the access request. See also static connection.

dynamic link libraries (DLLs) — Repositories of reusable software modules organized for dynamic linking; also refers to a Windows file format for storing reusable software modules.

dynamic linking — A linking process performed during program loading or execution. See also link editor.

dynamic RAM (DRAM) — A type of RAM stores each bit by using a single transistor and capacitor.

early bindingSee static linking.

effective data transfer rate — The data-transmission capacity actually achieved with a communication protocol; always less than the raw data transfer rate. See also raw data transfer rate.

electromagnetic interference (EMI) — An alteration of wave characteristics caused by external electrical or magnetic phenomena, such as radio equipment and nearby power lines.

electronically erasable programmable ROM (EEPROM) — A type of nonvolatile memory that can be programmed, erased, and reprogrammed by signals sent from a CPU; the only type of ROM that�s currently used.

encapsulation — A message-translation process that embeds all or part of a datagram in the message format of a physical network.

Enterprise JavaBeans (EJBs) — Java components that run in a business container on a server; capable of performing complex, behind-the-scenes processing.

erasable programmable ROM (EPROM) — A type of nonvolatile memory that�s manufactured blank, written (programmed) with a special EPROM writer, and erased by exposure to ultraviolet light.

Ethernet — A widely used LAN technology, developed by Xerox in the early 1970s, that�s closely related to the IEEE 802.3 standard. See also IEEE 802 standards and local area network (LAN).

even parity — An error-detection method in which the sender sets the parity bit to 0 if the count of 1-valued data bits is even or to 1 if the count of 1-valued data bits is odd. See also parity bit.

excess notation — A format that can be used to represent signed integers with a fixed number of bits; essentially, it divides a range of ordinary binary numbers in half and uses the lower half for negative values and the upper half for nonnegative values. See also signed integer.

exclusive OR (XOR) — An instruction that generates the value true if either (but not both) data input is true.

executable code — A program consisting entirely of CPU instructions that are ready to be loaded and run.

executing — The act of a processor performing a function in response to an instruction. See also instruction and processor.

execution cycle — The CPU cycle in which instructions are retrieved from registers, the specified data transformation is performed, and data outputs are stored in registers.

explicit priority — A priority-based scheduling method that assigns a priority level to each thread and can dispatch the highest-priority threads first or assign larger time slices to high-priority threads. See also dispatching, priority-based scheduling, and thread.

Extended Binary Coded Decimal Interchange Code (EBCDIC) — An IBM mainframe coding method for representing character data in an 8-bit format.

Extensible Markup Language (XML) — An extension of HTML that describes the structure, format, and content of documents. See also Hypertext Markup Language (HTML).

external function call — A placeholder in object code that the compiler generates for missing executable code; contains the name and type of the called function as well as the memory addresses and types of function parameters. See also compiler and function.

external I/O buses — Subsidiary buses that connect external devices to the system bus; they provide the connection points for external devices and aggregate their capacity to better match a system bus connection�s capacity.

fault tolerance — In an FMS, methods of securing file content against hardware failure. See also file management system (FMS).

fetch cycle — The CPU cycle in which data inputs are prepared for transformation into data outputs; instructions are fetched from primary storage and stored in registers.

fiber-optic cable — A guided transmission medium for optical signals, generally consisting of plastic or glass fibers sheathed in a protective plastic coating.

fields — In a file�s logical structure, the components of a record; usually contains information about a single person, a thing, or an event. See also record.

fifth-generation language (5GL) A nonprocedural programming language used to develop software that mimics human intelligence.

file — A sequence of records on secondary storage; the common organization schemes for files are sequential and indexed. See also record.

File Allocation Table (FAT) — The file system used in DOS and early Windows versions to record the allocation of storage device locations to files and directories.

file association — The relationship between file types and the programs or OS utilities that manipulate them.

file close operation — The act of severing the relationship between a file and a process by flushing file I/O buffers, deallocating buffer memory, updating the file�s directory entries, and updating the open file table.

file control layer — The FMS layer that provides service functions for manipulating files and directories, processes service calls from the command layer, issues commands to the storage I/O control layer to interact with hardware, and maintains the directory and storage allocation data structures. See also file management system (FMS).

file management system (FMS) — The collection of system software that performs file and secondary storage management and access functions; usually part of the OS.

file migration — A file management technique that balances each file version�s storage cost with anticipated user demand for this version; older versions of files are moved automatically to less costly storage media or devices.

file open operation — The process of associating a file with an active process by allocating buffers and updating internal tables.

File Transfer Protocol (FTP) — An older Web protocol that specifies a client/server request and response language for copying files from one Internet host to another.

firewall — A hardware device or software (or a combination) that prevents unauthorized users in one network from accessing resources on another network.

firmware — Software, such as system BIOS, stored in nonvolatile memory; can be loaded into main memory at high speeds. See also nonvolatile memory (NVM).

first come, first served (FCFS) — A priority-based scheduling method in which the scheduler always dispatches the ready thread that has been waiting the longest. See also dispatching, priority-based scheduling, and thread.

first-generation languages (1GLs) — The earliest programming languages, consisting of binary CPU instructions.

fixed-length instructions — In this type of instruction format, the amount by which the instruction pointer must be incremented after each fetch is a constant; this increment is the length of an instruction.

flag — A Boolean variable representing each bit in a PSW register�s bit string; generally used to store the result of a comparison operation, control conditional BRANCH execution, or indicate actual or potential error conditions. See also program status word (PSW).

flash RAM — The most common type of nonvolatile memory; typically used to store firmware and in portable secondary storage systems, such as USB flash drives.

flat memory model — An approach to assigning memory addresses in which memory locations are described by single unsigned integers corresponding to linear positions.

flat panel displays — Newer video display devices that are thinner, generate higher quality images, and consume less power than CRTs.

floating-point notation — A method for representing real numbers that consists of two parts: a mantissa and an exponent; the mantissa holds the bits that are interpreted to derive the real number�s digits, and the exponent value indicates the radix point�s position.. See also real number.

font — A collection of characters of similar style and appearance.

formula — A sequence of computation and data movement instructions that a processor executes to solve a processing problem.

forwarding table � A table of node addresses and transmission lines or connection ports that each central network node maintains to make message-forwarding decisions.

fourth-generation languages (4GLs) � High-level programming languages that have higher instruction explosion and support nonprocedural programming, database manipulation, and advanced I/O capabilities. See also instruction explosion.

fragmentation — The scattering of storage locations allocated to a single process or purpose throughout non-contiguous locations in physical memory or a secondary storage device. See also contiguous and memory allocation.

fragmented — The condition of a hard disk (or other storage drive) with many programs and files scattered across it in noncontiguous storage locations. See also contiguous.

frequency — The number of wave cycles occurring in 1 second; measured in hertz. See also cycle.

frequency-division multiplexing (FDM) — A channel-sharing technique that partitions a single broadband channel into multiple narrowband subchannels, each representing a different frequency band.

frequency modulation (FM) — A modulation method that represents bit values by varying carrier wave frequency while holding amplitude constant.

frequency-shift keying (FSK) See frequency modulation.

front-end CASE tool — An application development tool that primarily supports developing system models. See also computer-assisted software engineering (CASE) tool.

full backup — A type of backup in which the FMS copies all files and directories for an entire storage volume; can include storage allocation tables, partition tables, and other important disk management data structures.

full-duplex mode — A communication mode in which two transmission lines are used; allows simultaneous communication in both directions.

fully qualified referenceSee complete path.

function — A named instruction sequence in a high-level programming language that�s always executed as a unit; also called a subroutine or a procedure.

gate — A circuit that can perform a processing function on a single binary electrical signal, or bit. See also switch.

gateway — Node connecting two or more networks or network segments that might be physically implemented as workstations, servers, or routers.

general-purpose processor — A processor that can be instructed to perform a wide variety of tasks. See also processor.

general-purpose registers — Registers used only by the currently running program; they typically hold intermediate results or frequently used data values. See also registers.

germanium, antimony, and tellurium (GST) — A glasslike compound, used in phase-change memory, that can change between amorphous and crystalline states. See also phase-change memory (PCM).

Gigabit Ethernet — An Ethernet standard, based on the IEEE 802.3z and 802.3ab standards, for 1 Gbps transmission. See also Ethernet and IEEE 802 standards.

gigahertz (GHz) — A measurement of wave or system clock frequency; one billion cycles per second.

Google File System (GFS) — A file system Google developed to meet rapidly increasing storage requirements; offers scalability to petabyte storage, the capability to handle extremely large files, data storage on distributed general-purpose servers, and simultaneous file access by multiple distributed applications.

grandparent — The file version before the parent version. See also parent.

graph directory structure — A directory structure in which files and subdirectories can be contained in multiple directories, and directory links can form a cycle. See also links.

Graphics Interchange Format (GIF) — A common bitmap compression format for still images.

grayscale — A display device or encoding method that can display black, white, and shades of gray but no other colors.

grid — A group of dissimilar computers, connected by a high-speed network, that cooperate to provide services or run a shared application; unlike a cluster, computers in a grid work cooperatively at some times and independently at others and can be located far away from each other.

Grosch�s Law — An outdated mathematical relationship between computer size and cost per unit of instruction execution, which states that computing power, measured by millions of instructions per second, is proportional to the square of hardware cost. In other words, large and powerful computers are always more cost effective than smaller ones.

guided transmission — A transmission medium that routes signals between two locations through a physical connection, such as copper wire or optical fiber; also called wired transmission.

H.323 — The oldest and most widely deployed VoIP protocol suite; also addresses video and data conferencing. See also Voice over IP (VoIP).

half-duplex mode — A communication mode that uses a single shared channel, and each node takes turns using the transmission line to transmit and receive.

half-toning Simulating shades of gray by dithering black and white dots. See also dithering.

HALT — An instruction that suspends the normal flow of instruction execution in the current program; in some CPUs, it causes the CPU to cease all operations, and in others, it causes a BRANCH to a predetermined memory address. See also BRANCH.

hard disk — A magnetic disk medium with a rigid metal base (substrate) where data is recorded as patterns of magnetic charge.

hardware independence — Embedding hardware�s physical details into system software so that users and application programmers don�t need to know them to interact with hardware.

hardware monitor — A device attached directly to the communication link between two hardware devices, often used to monitor the use of communication channels, disk drives, and network traffic; monitors communication activity between the two devices and stores communication statistics that can be retrieved and printed in a report.

head-to-head (HTH) switching time — The time needed to switch a hard drive�s read/write circuitry to the correct read/write head before accessing a sector.

heat dissipation — Conducting excessive heat away from a device, thus reducing its temperature.

heat sink — An object designed to absorb heat and rapidly dissipate it via air or water movement; it�s placed in direct physical contact with an electrical device.

helical scanning — A geometric approach to recording data on a tape surface in which data is read and written by rotating the read/write head at an angle and moving from tape edge to tape edge. See also linear recording.

hertz (Hz) — In computers, a unit of measure for the frequency of system clock timing pulses; one Hz corresponds to one clock cycle per second. See also clock rate and system clock.

hexadecimal notation — A numbering system with a base value of 16; uses digits from 0 to 9 and letters from A to F, which represent the decimal values 10 to 15.

hierarchical directory structure — A multilevel system of directories in which directories can contain other directories, but a directory can�t be contained in more than one parent.

high-order bitSee most significant digit.

hit ratio — The ratio of cache hits to read accesses. See also cache hit.

home directory — The main directory associated with and owned by a single user.

host channel adapter (HCA) — An interface that connects a device to an InfiniBand switch; used by devices, such as general-purpose servers, that can initiate and respond to data transfer requests.

HTTPS — A secure version of HTTP that encrypts HTTP requests and responses. See also Hypertext Transfer Protocol (HTTP).

hub — A central connection point for nodes in a LAN; provides separate point-to-point connections between nodes and the hub by using cabling in a physical star topology and attaching these connections to its internal shared bus. See also local area network (LAN).

Hypertext Markup Language (HTML) — A device-independent formatting language that describes Web documents; links to other documents can be embedded in it.

Hypertext Transfer Protocol (HTTP) — A Web protocol that specifies the language by which clients request documents and how servers respond to those requests.

hypervisor — An OS that enables dividing a single physical computer or cluster into multiple virtual machines. See also virtual machine (VM).

IEEE 802 standards — A collection of IEEE network standards describing physical network hardware, transmission media, transmission methods, and protocols..

image description language (IDL) — A language (usually device independent) that uses compact bit strings or ordinary ASCII or Unicode text to describe primitive image components, such as straight lines and simple shapes; reduces storage space requirements because a description of a simple image component is usually much smaller than a bitmap.

implementation discipline — Activities in the Unified Process for building, acquiring, and integrating application software components. See also Unified Process (UP).

inclusive OR — An instruction that generates the value true if either or both data inputs are true; usually called just an �OR instruction.�

incremental backup — A type of backup in which the FMS archives only files that have been modified since the previous incremental or full backup. See also full backup.

index — In file organization, an array of pointers to records. See also pointer and record.

indirect addressing — A method of computing physical memory addresses automatically; the CPU adds the process offset to all memory address operands before accessing memory. See also process offset.

InfiniBand — A data connection standard for high-speed interconnection of network switches, servers, and secondary storage devices; based on a switched fabric architecture. See also switched fabric.

information architecture — The requirements and constraints that define important characteristics of information-processing resources and how these resources interact with one another.

infrastructure — Capital resources that provide benefits to a wide range of organizational units and functions; typically, they serve a large, diverse group of users; have high maintenance costs; involve costs difficult to allocate to users separately; and require recurring capital expenditures. See also capital resources.

infrastructure as a service (IaaS) — A cloud-based architectural approach similar to PaaS, in which customers can configure application and system software for a generic platform as virtual servers and then deploy these servers to a third-party hosting site; often used to provide back-end storage services and large-scale computing infrastructures for running complex simulations. See also platform as a service (PaaS).

inkjet printer — A printer that produces printed images by placing small drops of liquid ink onto paper; ink is forced out of the nozzle by mechanical movement or by heat.

input/output (I/O) units — Devices that perform external communication functions.

input pads — A general class of input devices that convert pressure into input, such as for capturing signatures or drawings; these devices typically use infrared detectors, photosensors, pressure-sensitive pads, or magnetic fields.

Institute for Electrical and Electronics Engineers (IEEE) Computer Society — A subgroup of the IEEE that specializes in computer and data communication technologies.

instruction — A signal or command to a processor to perform one of its functions. See also processor.

instruction cycleSee fetch cycle.

instruction explosion — The one-to-many (1:N) relationship between later-generation programming statements and the CPU actions implementing them.

instruction format — A template that specifies the number of operands and the position and length of the op code and operands in instructions.

instruction pointer (IP) — A special-purpose register that stores the address of the next instruction the control unit should fetch from memory. See also registers.

instruction register — A special-purpose register that holds an instruction the control unit has fetched from memory. See also registers.

instruction set — The collection of instructions that a CPU can process. See also instruction.

integer — A whole number�that is, a value that doesn�t have a fractional part.

integrated circuit (IC) — A semiconductor device that incorporates several transistors and their interconnections on a single chip.

integrated development environment (IDE) — A collection of automated support tools to speed program development and testing; typically includes tools such as program editors, interpreters, compilers, debuggers, prototyping tools, function and class libraries, and so forth.

interleaved executionSee concurrent execution.

International Alphabet 5 (IA5) — The international equivalent of the ASCII coding method for character data. See also American Standard Code for Information Interchange (ASCII).

International Organization for Standardization (ISO) — An international group with functions similar to those of the American National Standards Institute.

Internet — A global collection of networks interconnected with TCP/IP.

Internet Inter-ORB Protocol (IIOP) A component message-passing protocol in CORBA. See also Common Object Request Broker Architecture (CORBA).

Internet Message Access Protocol 4 (IMAP4) — A protocol that extends POP3 to permanently store and manage e-mail messages on the server, which enables users to access stored e-mail from any Internet host. See also Post Office Protocol 3 (POP3).

Internet Protocol (IP) — A core protocol for packet switching and routing on the Internet.

Internet Protocol version 4 (IPv4) — The original version of IP; uses 32-bit addresses.

Internet Protocol version 6 (IPv6) — An update to IPv4 that uses 128-bit addresses; developed to address the limited number node addresses, handle streaming multimedia, and multicast more efficiently.

interpretation — A process for source code instructions that interleaves source code translation, link editing, and execution. See also link editor and source code.

interpreter — A program that reads a single source code instruction, translates it into CPU instructions or a DLL call, and executes the instructions or DLL call immediately. See also dynamic link libraries (DLLs).

interrupt — A signal sent to the CPU over the control bus that some event requires it to execute a specific program or process; used to prevent inefficiency caused by I/O wait states. See also I/O wait states.

interrupt code — A numerical value of an interrupt, indicating the type of event that has occurred; usually equivalent to the bus port number of the peripheral device sending the interrupt. See also interrupt.

interrupt handler — An OS service routine that processes interrupts; each interrupt handler is a separate program stored in a separate part of primary storage.

interrupt register — A register in the CPU�s control unit that stores interrupt codes received over the bus or generated by the CPU. See also interrupt code.

intranet — An internal private network that uses Internet protocols but is accessible only by a limited set of internal users; also describes privately accessible resources that are organized and delivered via one or more Web protocols over a TCP/IP network.

I/O channel — A device controller dedicated to a mainframe bus port that enables many devices to share access to, and the capacity of, the port; originally used to describe a specific hardware component of IBM�s 7000 series mainframe computers.

I/O port — A communication pathway from the CPU to a peripheral device; in most computers, it�s a memory address, or set of contiguous memory addresses, that can be read or written by the CPU and a single peripheral device.

I/O wait states — Idle processor cycles consumed while waiting for secondary storage or I/O devices to complete access requests.

iterations — Repeated steps in an SDLC process; for example, in the UP, each iteration, consisting of specific activities, is 4 to 6 weeks. See also systems development life cycle (SDLC) and Unified Process (UP).

Java — An object-oriented programming language that supports almost any combination of hardware platform and OS.

Java Platform, Enterprise Edition (Java EE) A family of standards for developing and deploying component-based distributed applications written in Java; follows a three-layer architecture, with the client, Web/business, and data tiers.

JavaServer Faces (JSF) Java components that enable developers to create user interfaces that run on a server but interact with a client Web browser or component.

JavaServer Pages (JSP) — Server-side Java components that generate formatted Web pages by using embedded scripts.

Java Virtual Machine (JVM) — A hypothetical computer and operating system serving as the target machine for Java interpreters and compilers; avoids the need to translate source code instructions for a specific platform and OS.

job control language (JCL) See command language.

Joint Photographic Experts Group (JPEG) — A common bitmap compression format for still images.

journalingSee transaction logging.

JUMPSee BRANCH.

keyboard controller — A microprocessor integrated into a keyboard that generates a bitstream output of scan codes according to an internal program or lookup table. See also scan code.

Kerberos — A security model that defines standard interactions between clients, services, and a trusted security service.

kernel — The OS layer that manages resources and interacts with hardware; includes a resource allocation layer and interface programs called device drivers.

label — A mnemonic representing a program instruction�s memory address.

large-format printer — A more current term for plotters. See also plotter.

laptop computer — A full-featured, portable microcomputer with an integrated display and a battery; rivals traditional microcomputers in power and cost. See also microcomputer.

laser printer — A printer that operates by charging areas of a photoconductive drum; toner is attracted to charged areas of the drum and then to paper.

late bindingSee dynamic linking.

Latin-1 — An ISO standard character-coding table containing the ASCII-7 characters in the lower 128 table entries and most of the characters used by Western European languages in the upper 128 table entries. See also multinational characters.

law of diminishing returns — The economic principle stating that when multiple resources are required to produce something useful, adding more of a single resource produces fewer benefits; can be applied to buffer and cache sizes as well as many other computer system components.

least significant byte — In storage bytes, the rightmost byte in a multiple-byte data item containing digits of the lowest weight.

least significant digit — The rightmost digit in a bit string that represents the lowest weight.

level one (L1) cache — An SRAM cache between the CPU and SDRAM primary storage, used to limit wait states; the L1 cache is closest to the CPU.

level two (L2) cache — An SRAM cache between the CPU and SDRAM primary storage, used to limit wait states; the L2 cache is the next level away from the CPU.

level three (L3) cache — An SRAM cache between the CPU and SDRAM primary storage, used to limit wait states; the L3 cache is the farthest level away from the CPU.

light-emitting diode  (LED) — A video display technology that uses phosphorescent compounds to produce red, green, and blue light.

Lightweight Directory Access Protocol (LDAP) — An Internet standard for directory services, based on the X.500 standard and adopted by the Internet Engineering Task Force.

linear address space — The set of sequentially numbered storage locations in a peripheral device; these locations must be converted into a disk�s corresponding platter, sector, and track for the CPU to physically access the correct sector. See also logical access.

linear recording — A geometric approach to recording data on a tape surface in which bits are placed along parallel tracks that run along the tape�s entire length. See also helical scanning.

Linear Tape Open (LTO) — A magnetic tape standard developed by Hewlett-Packard, IBM, and Seagate; uses linear recording and has technology improvements in tape cartridges, coercible materials, read/write heads, and tape control. See also linear recording.

line turnaround — A control message that�s sent when one node in a half-duplex channel has stopped sending; the receiver then assumes the role of sender.

links — The UNIX term for pointers from one directory to another in a graph directory structure; also refers to pointers connecting two data items in a data structure or external function calls in an object code file. See also graph directory structure.

link editor — A program that combines object code files into an integrated set of executable code with a consistent scheme of memory addresses and references.

linked list — A data structure that uses pointers so that list elements can be scattered among nonsequential storage locations. See also doubly linked list and singly linked list.

liquid crystal display (LCD) — A video display device containing liquid crystals sandwiched between two polarizing filter panels; the crystals change from opaque to transparent when an electrical charge is applied.

little endian — A CPU or memory architecture in which the least significant byte is stored at the lowest memory address. See also least significant byte.

load — A data transfer from main memory into a register.

local area network (LAN) — A network that spans a limited area, such as a single building or office floor.

location transparency — A characteristic of software and user interfaces meaning that local and remote resources are accessed in the same way; also referred to as �network transparency.�

logic instructionsInstructions that implement Boolean operations, such as ADD, AND, OR, NOR, and XOR.

logical access — A read or write operation from the hypothetical storage device representing a peripheral device. See also linear address space.

logical record — A a collection of data items, or fields, that an application program accesses as a single unit.

logical SHIFT — A SHIFT instruction used to extract a single bit from a bit string. See also SHIFT.

logical topology — The path messages traverse as they travel between end and central network nodes. See also network topology.

long integers — Double-precision representations of integers. See also double-precision.

longitudinal redundancy checking (LRC) See block checking.

lossless compression — A compression algorithm in which data content is unchanged when compressed and then decompressed.

lossy compression — A compression algorithm in which data content is altered or lost when compressed and then decompressed; usually applied only to audio and video data.

low-order bit See least significant digit.

machine data typesSee primitive data types.

machine independenceSee hardware independence.

machine languagesSee first-generation languages (1GLs).

machine state — The saved register values of interrupted processes or programs that represent their state before an interrupt. See also interrupt.

magnetic decay — The tendency of magnetically charged particles to lose their charge over time; it�s constant over time and proportional to the power of the charge.

magnetic leakage — The reduction in strength of a stored magnetic charge because of interference from adjacent magnetic charges of opposite polarity.

magnetic tape — A ribbon of plastic with a coercible (usually metallic oxide) coating, used to store data.

magneto-optical (MO) drive — A secondary storage device that uses a laser and reflected light to sense magnetically recorded bit values; data reading is based on the polarity of the reflected laser light, which is determined by the polarity of the magnetic charge.

magnetoresistive RAM (MRAM) — A type of nonvolatile memory under development that stores bit values by using two magnetic elements, one with fixed polarity and the other with polarity that changes when a bit is written; has better longevity than conventional flash RAM.

main memorySee primary storage.

mainframe — A computer system designed to handle the information-processing needs of a large number of users and applications and optimized to store large quantities of data and move it from one place to another quickly and efficiently.

Mammoth — A magnetic tape standard developed by Exabyte based on Digital Audio Tape; uses helical scanning and an improved tape drive technology to pack more data onto a single tape. See also Digital Audio Tape (DAT) and helical scanning.

manipulation — In computer processing, refers to manipulating data  by executing processor instructions, such as addition, subtraction, and equality comparisons.

mark sensor — An optical input device that scans for light or dark marks at specific locations on a page.

Media Access Control (MAC) — A protocol for determining how to share a transmission medium efficiently.

megahertz (MHz) — A measurement of wave or system clock frequency; one million cycles per second.

memory allocation — The assignment of specific memory addresses to system software, application programs, threads and processes, and data.

memory bus — A subsidiary bus that connects only the CPU and memory; because its data transfer rate is higher than that of the system bus, it improves overall computer performance.

memory map — A list of the memory location of every function and program variable; produced by the link editor based on the symbol table�s contents. See also link editor and symbol table.

memory protection — A procedure for protecting memory allocated to one program from unauthorized access by another program; adds overhead to each write operation.

mesh topology — A network topology in which every node pair is connected by a point-to-point link; requires many transmission lines if the number of end nodes is large. See also network topology.

message — In network communication, a unit of data or information transmitted from a sender to a recipient. In components and object-oriented programs, a request sent from one object or component to another.

methods — Programs for manipulating data items in a class. See also class.

metropolitan area networks (MANs) — Networks that typically cover a town or city; the market for WiMAX networks. See also Worldwide Interoperability for Microwave Access (WiMAX).

microchip — A semiconductor device capable of integrating hundreds, thousands, and even billions of electrical devices on a single chip.

microcomputer — Also called a PC or a workstation, a computer system designed to meet a single user�s information-processing needs; can include portable computers, such as laptops and handheld computers.

microprocessor — A microchip containing all the circuits and components of a CPU.

middleware — System software that that �glues� together parts of a client/server or multitier application and enables clients and servers or distributed components to locate and communicate with one another.

midrange computer — A computer system designed to provide information processing for multiple users and run many application programs simultaneously; sometimes called a �minicomputer.�

millions of floating-point operations per second (megaflops or MFLOPS) — A measurement of the rate at which floating-point operations are performed; used to measure CPU performance.

millions of instructions per second (MIPS) — A measurement of the rate at which instructions are executed; assumed to measure CPU performance when manipulating single-precision integers.

modulator-demodulator (modem) A device that translates analog signals into digital signals and vice versa, enabling computer hardware to communicate over voice-grade phone lines.

monitors Video display panels. The term also refers to hardware or software that tracks and reports processing or I/O activity. See also hardware monitor and software monitor.

monochrome — A display device or encoding method that can display only one of two colors (usually black and white), so it requires only 1 bit per pixel.

monophonic — Capable of generating only one frequency (note) at a time.

Moore�s Law — Gordon Moore�s observation that the rate of increase in transistor density on microchips doubles every 18 to 24 months, with no increase in unit cost.

most significant byte — In storage bytes, the leftmost byte in a multiple-byte data item containing digits of the highest weight.

most significant digit — The leftmost digit in a bit string that represents the greatest weight.

MOVE — An instruction that copies data bits to storage locations and can copy data between any combination of registers and primary storage locations.

Moving Pictures Experts Group (MPEG) — An organization that creates standards for motion picture recording and encoding technology; each standard is divided into layers numbered 1 (systems), 2 (video), and 3 (audio).

MP3 — The audio-encoding standard that�s layer 3 of the MPEG-1 standard. See also Moving Pictures Experts Group (MPEG).

multicasting — Transmission situations involving multiple senders and receivers.

multicomputer configuration — Any arrangement of multiple computers used to support specific services or applications; includes clusters, blades, and grids.

multicore architecture — A microprocessor architecture that embeds multiple CPUs and cache memory on a single chip.

multilevel coding — A technique for embedding multiple bit values in a single wave characteristic, such as frequency or amplitude; treats groups of bits as a single unit for the purpose of signal encoding.

multimode graded-index cable — A multimode fiber-optic cable in which fibers vary in density from the center to the edge, which reduces the number of light reflections.

multimode step-index cable — A multimode fiber-optic cable in which both the optical fiber and cladding have different but uniform densities throughout the cable, resulting in many light reflections.

multinational characters — Modified Latin characters, such as � and �, used by Western European languages other than English.

multiple-core CPUs — Processors that improve performance by integrating multiple processing cores and memory caches on a single chip and by increasing raw CPU speed; the processors share primary storage and a single system bus.

multiple-processor architecture — A more traditional approach to multiprocessing that uses two or more processors on a single motherboard or set of interconnected motherboards; slower than multicore architecture. See also multicore architecture and multiprocessing.

multiprocessing — Any CPU architecture in which duplicate CPUs or processor stages can execute in parallel; a form of parallel processing.

Multipurpose Internet Mail Extensions (MIME) — A protocol that�s an extension of SMTP; enables including nontext content in e-mail messages. See also Simple Mail Transfer Protocol (SMTP).

multitasking — An operating system�s support for running multiple programs simultaneously.

multithreaded — A process or program divided into two or more threads, each of which can be scheduled and executed independently. See also thread.

Musical Instrument Digital Interface (MIDI) — A standard for storing and transporting control information between computers and electronic musical instruments and synthesizers.

n-layer architecture — A client/server architecture with more than three layers; used when processing requirements or data resources are complex.

n-tiered architectureSee n-layer architecture.

named pipe — A pipe with a name that�s placed permanently in a file system directory; can communicate between processes on different computers. See also pipe.

narrowband — A low-bandwidth communication channel; typically a subchannel of a broadband channel.

native applications — Progams that are compiled and linked for a particular CPU and OS.

negative acknowledge (NAK) — An ASCII control character sent by a receiver if data errors are detected.

netbook computer — A laptop computer that emphasizes small size, reduced weight, low cost, and wireless networking; capable of performing only light-duty tasks. See also laptop computer.

network adapterSee network interface card (NIC).

network administrator — The person who�s responsible for managing an organization�s network infrastructure; can also apply to the manager of a local area network.

network-attached storage (NAS) — A storage architecture with a dedicated storage server attached to a general-purpose network to handle storage access requests from other servers.

network interface card (NIC) — A device that connects a node, such as a computer or network printer, to a network transmission cable.

Network layer — The OSI layer that forwards messages to their correct destinations. See also Open Systems Interconnection (OSI) model.

network topology — The spatial organization of network devices, physical routing of network cabling, and flow of messages from one network node to another.

network transparencySee location transparency.

New Technology File System (NTFS) — The Microsoft file system introduced with Windows NT; targeted to applications requiring high-speed directory and file operations, reliability and fault tolerance, secure file and disk content, and the capability to handle large disks, files, and directories.

noise — In a communication channel, unwanted signal components added to the data signal that might be interpreted incorrectly as data; can be introduced by factors such as electromagnetic interference and distortion.

noncontiguous memory allocation — A memory allocation scheme in which portions of a process can be allocated to free partitions anywhere in memory; uses small fixed-size partitions. See also contiguous and memory allocation.

nonprocedural language — A programming language that describes a processing requirement without describing a specific procedure for satisfying the requirement.

nonvolatile — A term describing storage devices that hold data without loss for long periods; secondary storage is usually nonvolatile.

nonvolatile memory (NVM) — A generic term for memory devices with long-term or permanent data retention.

NOT — An instruction that transforms the Boolean value true (1) into false (0) and the value false into true.

numeric range The set of all data values that can be represented by a data-encoding method.

object — One instance, or variable, of a class. See also class.

object code — The output of an assembler or a compiler; contains a mixture of CPU instructions, library calls, and other information the link editor needs. See also link editor.

object-oriented programming (OOP) — A programming paradigm that views data and programs as two parts of an integrated whole; better addresses program reuse and long-term software maintenance.

Object Request Broker (ORB) — A CORBA service that maintains a component directory and routes messages between components. See also Common Object Request Broker Architecture (CORBA).

objectclass — An LDAP concept that defines attributes common to all members of a class. See also directory services and Lightweight Directory Access Protocol (LDAP).

octal notation — A base-8 numbering system that uses digits from 0 to 7.

odd parity — An error-detection method in which the sender sets the parity bit to 0 if the count of 1-valued data bits in the character is odd and to 1 if the count of 1-valued data bits is even. See also parity bit.

offset register — A register containing the process offset value; used in indirect addressing. See also indirect addressing and process offset.

on-off keying (OOK) — A signal-coding method that generates square waves by rapidly switching (pulsing) an electrical or optical power source to represent bit values; essentially the digital equivalent of amplitude modulation. See also amplitude modulation (AM).

op code — In an instruction�s bit string, it�s the first group of bits and represents the instruction�s unique binary number.

OpenGL — A video controller image description language developed by Silicon Graphics but now maintained by Khronos Group as an open standard.

Open Systems Interconnection (OSI) model — A conceptual model for network hardware and software that organizes network functions into seven layers; useful as a general model of networks, a framework for comparing networks, and an architectural roadmap that enhances interoperability between network architectures and products.

operands — In an instruction�s bit string, they�re groups of bits after the op code that hold the instruction�s input values; they can contain a data item or the location of a data item. See also op code.

operating expenditures — Funds expended during the current fiscal year to support normal business activities.

operating system (OS) — A collection of utility programs for supporting users and application programs, allocating resources to multiple users and application programs, and controlling access to hardware.

optical character recognition (OCR) — A technology that combines optical-scanning technology with a special-purpose processor or software to interpret bitmap content. See also optical scanner.

optical scanner — A device that generates bitmap representations of printed images; detects light reflected from the page with an array of photosensors.

organic LED (OLED) — A newer LED, manufactured with TFT technology, that achieves high-quality color display with organic compounds. See also light-emitting diodes (LEDs) and thin film transistor (TFT).

overflow — An error that occurs when the result of a processing operation exceeds the format�s numeric range. See also numeric range.

packets — Basic units of data communication in a network.

packet switching — The most common type of TDM, in which messages are divided into packets and then transmitted to their destination as channel capacity becomes available. See also time-division multiplexing (TDM).

packet-filtering firewall — The simplest type of firewall; examines each packet and matches header content to a list of allowed or denied packet types.

page — A small fixed-size portion of a program, normally between 1 and 4 KB, swapped between primary and secondary storage. See also virtual memory management.

page fault — A reference to a page held in secondary storage. See also page and virtual memory management.

page fileSee swap space.

page frame — A memory page used in virtual memory management. See also page and virtual memory management.

page hit — A reference to a page held in memory. See also page and virtual memory management.

page tables — Tables that store information about page locations, allocated page frames, and secondary storage space. See also page and virtual memory management.

palette — A table of colors used to represent pixel color; the number of bits used to represent each pixel determines the table size.

parallel access — An access method that can access multiple storage locations simultaneously; can also be achieved by subdividing data items and storing the component pieces on multiple storage devices.

parallel transmission — Sending each bit position of a message over a separate transmission line simultaneously.

parent — The original version of a file, after updates have been applied to generate a new version (the child). See also child.

parent process — The original process that initiates and controls execution of a child process. See also process and spawn.

parity bit — A bit appended to a character that stores redundant information used for error checking; its value is a count of other bit values in the character.

parity checking — Validating character data by recomputing the value of a parity bit. See also even parity and odd parity.

passive matrix display — An LCD that shares transistors among rows and columns of pixels.

peer-to-peer (P2P) architecture — A sofware architecture in which the roles of client and server are combined into a single application or group of related applications.

peer-to-peer bus — In this arrangement, any device can assume control of the bus or act as a bus master for transfers to any other device. See also bus arbitration unit and bus master.

Pentium processors — A family of processors that improved memory access and raw CPU speeds and added features such as support for higher-speed system buses, pipelined instruction execution, and multimedia processing instructions.

Peripheral Component Interconnect (PCI) — A family of bus standards, developed by Intel in the early 1990s, found in nearly all small and midrange computers of that era; largely replaced by PCI Express since the late 2000s.

peripheral devices — Storage and I/O devices in a computer, other than the CPU and primary storage.

personal computer (PC) See microcomputer.

personal digital assistant (PDA) — A handheld computer, usually integrated with a cell phone, that supports light-duty tasks.

petaflops (PFLOPS) — A measurement of the rate (1015 per second) at which floating-point operations are performed; used to measure CPU performance.

phase — In communication, a specific time point in a sine wave�s cycle; measured in degrees. See also cycle.

phase-change memory (PCM) — A type of nonvolatile memory under development that uses a GST compound capable of switching between amorphous and crystalline states; has fast write times and high longevity. See also germanium, antimony, and tellurium (GST).

phase-shift keying (PSK) See phase-shift modulation.

phase-shift modulation — A modulation method that represents bit values as sudden shifts in wave phase.

phonemes — Vocal sounds that are basic components of human speech; they correspond roughly to the sounds of each letter of the alphabet..

photosensor — A device that converts incoming light energy into outgoing electrical energy.

Physical layer — The OSI layer where communication between devices actually takes place; includes hardware devices that encode and decode bitstreams and the transmission lines that transport them. See also Open Systems Interconnection (OSI) model.

physical memory — The actual number of memory bytes that are physically installed in a computer system; can be smaller than addressable memory but never larger. See also addressable memory.

physical record — The unit of storage transferred between the device controller and memory in a single operation.

physical topology — The physical placement of cables and device connections in a network topology. See also network topology.

pipe — A region of shared memory through which multiple processes executing on the same computer can exchange data; used for communicating between OS components, queuing requests to an OS service, and exchanging messages between program components.

pipelining — A method of organizing CPU circuitry so that multiple instructions can be in different stages of execution at the same time; a form of parallel processing.

pixel — An abbreviation of �picture element,� it�s a single unit of data in an image; also refers to a single point on a display surface.

plasma display — A video display device that uses an active matrix display and generates light by applying an electrical charge to neon gas.

platform as a service (PaaS) — A cloud-based architectural approach in which an organization rents access to system software and hardware on which it installs its own application software and other services.

platters — In magnetic disk media, they�re flat, circular disks with metallic coatings that are rotated beneath read/write heads; data is normally recorded on both sides.

plotter — A printer that generates line drawings on sheets or rolls of paper up to 64 inches wide. See also large-format printer.

point — A standard measurement unit for font size; equals 1/72 of an inch.

pointer — A data element containing the address of another data element; typically used in data structures. This term is also a synonym for �cursor� in a display device.

polyphonic — Capable of generating many frequencies (notes) simultaneously.

pop — The process of removing register values from the top of a stack and loading them back into the correct registers. See also stack.

port — A TCP connection identified by a unique integer number; many ports are standardized to specific Internet services. See also Transmission Control Protocol (TCP).

Portable Document Format (PDF) — An Adobe image description language developed to generate and manage documents as an integrated whole rather than a collection of independent images and pages.

Post Office Protocol 3 (POP3) — A protocol that standardizes the interaction between e-mail clients and servers so that client and server can run on different Internet hosts; e-mails are held on the server temporarily, downloaded to the client when a connection is established, and then deleted from the server.

PostScript — An Adobe image description language designed mainly for printed documents, although it�s also a programming language for generating video display output.

power sags — Momentary reductions in the voltage or amperage of electrical power.

power surges — Momentary increases (spikes) in the voltage or amperage of electrical power.

preemptive scheduling — A scheduling method that enables a higher-priority thread to interrupt and suspend a lower-priority thread. See also scheduling and thread.

Presentation layer — The OSI layer that makes sure data transmitted by one network node is interpreted correctly by the other network node; used mainly by applications that format data for user display. See also Open Systems Interconnection (OSI) model.

primary storage — High-speed storage in a computer system that holds currently running programs and data immediately needed by these programs.

primitive data types — The integer, real number, character, Boolean, and memory address data types that CPUs can manipulate directly.

priority-based scheduling — A scheduling method that determines which ready thread should be dispatched to the CPU based on user or thread priority. See also scheduling and thread.

procedureSee function.

process — A unit of executing software that�s managed independently by the OS and can request and receive hardware resources and OS services.Also refers to tranforming input data by applying calculations, manipulations, and other operations.

process control block (PCB) — A data structure containing information about an active process; used by the OS to keep track of each process and allocate resources, secure resource access, and protect active processes from interference by other active processes.

process family — The collective name for a parent process and all its descendants. See also process and spawn.

process listSee process queue.

process offset — In memory allocation, the difference between the first address in physical memory and the address of the first process instruction. See also memory allocation and process.

processor — A device capable of performing data manipulation and transformation operations.

process queue — A data structure containing a list of PCBs for all active processes; can be searched by OS components. See also process control block (PCB).

production version — A program version that omits the symbol table and debugging checkpoints to reduce program size and increase execution speed. See also debugging version and symbol table.

program — A stored set of instructions for performing a specific task.

program editors — Writing tools similar to word-processing applications but customized for writing programs instead of documents. See also application development software.

program profiler — A software utility that describes the resource or service utilization of an application program during execution.

program status word (PSW) — A special-purpose register containing a bit string describing the CPU�s status and the currently running program. See also flag and registers.

program translator — A program that translates instructions in a programming language into CPU instructions. See also application development software.

programmer — A software developer who builds and tests software; might also perform tasks in the requirements and design disciplines. See also software developers.

programming language — A language for expressing computer-processing functions or instructions.

protocol stack — A complex set of software layers that an OS uses to implement network I/O and services.

proxy serverSee application firewall.

push — The process of copying register values to the top of a stack. See also stack.

Quarter Inch Committee (QIC) — A committee that develops open standards for magnetic tape drives on smaller computers.

qubit — An atom or any other matter that stores data in multiple simultaneous quantum states.

radio frequency (RF) — Electromagnetic radiation propagated through space; describes transmissions using frequencies between 50 Hz and 1 THz.

radixSee base.

radix point — In numbering systems other than decimal, the period or comma that separates the whole and fractional parts of a numeric value. See also decimal point.

random access — An access method that can access any storage location directly and in any order; primary storage devices and disk storage devices use random access.

random access memory (RAM) — Semiconductor devices used to implement primary storage; they don�t provide permanent storage because RAM�s contents are erased when the system power is turned off.

raw data transfer rate — The maximum number of bits or bytes per second a communication channel can carry; ignores the communication protocol and assumes error-free transmission.

read-only memory (ROM) — The earliest type of nonvolatile memory, with data content written permanently during manufacture; this primary storage device can be read, but no further data can be written.

read/write head — A mechanism in a storage device that reads and writes data to and from the storage medium; also referred to as a �read/write mechanism.�

ready state — The state of an active thread that�s idle, pending availability of a CPU. See also thread.

ready-to-send (RTS) signal — A CSMA/CA signal sent by a node waiting to transmit. See also Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA).

real number — A number that can contain both whole and fractional components; the fractional portion is represented by digits to the right of the radix point. See also radix point.

real resources — A computer�s hardware devices and associated system software that physically exist. See also virtual resources.

real-time scheduling — A scheduling method that guarantees a thread a minimum amount of CPU time and enough resources to complete its function in a specified time if the thread makes an explicit request when it�s created. See also scheduling and thread.

record — A data structure composed of other data structures or primitive data elements; commonly used as a unit of input and output to and from files or databases.

reduced instruction set computing (RISC) — A computer and processor design approach that typically includes fixed-length instructions, short instruction length, and a large number of general-purpose registers; the main feature is the absence of some complex instructions from the instruction set. See also complex instruction set computing (CISC).

redundant array of independent disks (RAID) — A disk storage technique that improves performance and fault tolerance. See also fault tolerance.

refresh cycle — In dynamic RAM, the period during which circuitry supplies fresh infusions of power automatically; read and write operations can�t be performed during this cycle. Also refers to the transfer of a full screen of data from the display generator to the monitor.

refresh rate The number of refresh cycles per second on a video display device; normally stated in hertz. See also refresh cycle.

registers — Internal storage locations in a CPU; each is capable of holding a single instruction or data item. See also central processing unit (CPU).

relative addressingSee indirect addressing.

relative path — The access path that begins at the current directory�s level and extends downward to a specific file. See also current directory.

release versionSee production version.

Remote Procedure Call (RPC) — A protocol that enables a process on one computer to call a process on another computer.

repeater — A device that functions much like an amplifier but extracts data embedded in the signal it receives and retransmits a new signal containing the same data; therefore, noise or distortion aren�t retransmitted. See also amplifier.

request for proposal (RFP) — A formal document stating hardware or software requirements and soliciting proposals from vendors to meet these requirements.

requirements discipline — Activities in the Unified Process for developing models of system and user requirements. See also Unified Process (UP).

resistance — The loss of electrical power that occurs as electrons pass through a conductor; low resistance means little power is lost.

resolution — The number of pixels displayed per linear measurement unit.

resource registry — A database, maintained by the resource locator, containing the names and locations of known resources and services on a network.

return instruction — In programming, an instruction executed at the end of a function to return control to the calling function. See also function.

return wire — In electrical transmission through wires, the channel component that completes an electrical circuit between sending and receiving devices.

RGB — Red, green, and blue; the primary colors for video display.

ring topology — A network topology in which every node node is connected directly to two other nodes, with a set of links forming a closed ring. See also network topology.

Rock�s Law — Arthur Rock�s addendum to Moore�s Law, stating that the cost of fabrication facilities for the latest chip generation doubles every four years. See also Moore�s Law.

rotational delay — The time a hard disk controller must wait for the right sector to rotate beneath read/write heads.

router — A device that intelligently forwards messages between networks; it stores messages in a buffer, examines their contents, and applies decision rules to determine where to forward them.

routing tables — Internal maps of a network containing information routers use to forward messages and choose from multiple possible paths to a recipient; routers periodically exchange this information with other routers to learn about networks beyond those to which they�re connected. See also router.

run queue — A data structure listing all active TCBs. See also thread and thread control block (TCB).

running state — The state of an active thread that�s been dispatched and has CPU control. See also dispatching and thread.

sampling — The process of converting analog sound waves to digital representation; it analyzes the content of the audio sound spectrum many times per second and converts it to a numeric representation.

sandbox — The protected area in which Java applets and servlets run; provides extensive security controls to prevent these programs from accessing unauthorized resources or damaging the hardware, OS, or file system. See also Java.

scaling out — An approach to increasing processing and other computer system power by partitioning processing and other tasks among multiple computer systems; examples include clusters and grids. See also cluster and grid.

scaling up — An approach to increasing processing and other computer system power by using larger and more powerful computers; examples include multicore and multiple-processor architectures. See also multicore architecture and multiple-processor architecture.

scan code — A 1- or 2-byte data element generated by a keyboard controller; represents a specific keyboard event. See also keyboard controller.

scanning lasers — Devices that sweep a narrow laser beam back and forth across bar codes. See also bar-code scanner.

scheduler — The portion of the OS that makes scheduling decisions for threads. See also thread.

scheduling — The decision-making process the OS uses to determine which ready thread moves to the running state. See also thread.

scripting language — A simple programming language that enables programmers to assemble software quickly by �gluing� together the capabilities of many other programs, such as Web servers and database management systems; scripts can be embedded in HTML pages and many other programs.

second-generation language (2GL) See assembly language.

secondary storage — System devices that provide large-capacity and long-term data storage.

sector — The data transfer unit for magnetic disk and optical disc drives; the size is generally stated in bytes and can vary from one device to another. Also refers to a fractional portion of a track on magnetic disk media.

Secure Shell (SSH) — An improved version of Telnet that encrypts data between client and server to address a major security issue in Telnet. See also Telnet.

segmented memory model — An approach to assigning memory addresses in which primary storage is divided into equal-sized segments called pages, identified by sequential nonnegative integers; each byte of memory has a two-part address: The first part identifies the page, and the second part identifies the byte in the page.

semiconductors — Materials with conductivity that varies in response to the electrical inputs applied; they have resistance properties that can be modified between those of a conductor and an insulator by adding chemical impurities.

sequential access time — The time required to read the second of two adjacent sectors on the same track and platter of a hard disk.

serial access — An access method that stores and retrieves data items in a linear (sequential) order; mainly used to hold backup copies of data stored on other storage devices.

Serial Advanced Technology Attachment (SATA) — A storage device and cabling standard commonly used in PCs; compatible with older parallel ATA standards but uses serial transmission.

Serial Attached SCSI (SAS) — A storage device and cabling standard commonly used in servers; compatible with older parallel SCSI standards but uses serial transmission.

serial transmission — Sending bits sequentially over a single transmission line; the receiver reassembles the bits into larger data units, such as bytes.

server — A mode of use rather than a class of computer system; manages shared resources and enables users to access these resources over a network.

server consolidation — Using virtual machines as small servers hosted by a hypervisor running on a larger machine; reduces total hardware requirements because not every server has to be installed on a separate computer. See also hypervisor and virtual machine (VM).

service call — A request to execute an OS service-layer function.

service layer — The OS layer containing reusable components packaged as functions that can be called from other programs; also acts as an intermediary between programs, which request and use resources, and the kernel, which manages and provides access to resources. See also kernel.

service-oriented architecture (SOA) — A design philosophy under which operating systems and some application software are constructed as a set of services that can be accessed by both internal and external users and software components.

service standards — Standards for providing infrastructure-based services to a wide variety of users. See also infrastructure.

servlet — A full-fledged Java program that runs in a Web server and performs functions such as calculations, database access, and creation of Web pages; runs in a protected area called the sandbox. See also Java Platform, Enterprise Edition (Java EE).

Session layer — The OSI layer that establishes and manages communication sessions. See also Open Systems Interconnection (OSI) model.

shellSee command layer.

SHIFT — An instruction that moves all bit values right or left, according to the number of positions specified by the operand; after shifting, empty positions are filled with 0s, and bit values that shift beyond the bit string�s bounds are discarded.

shortcuts — The Windows term for links in a graph directory structure. See also graph directory structure and links.

shortest time remaining (STR) — A priority-based scheduling method that chooses the next thread to be dispatched based on the expected amount of CPU time needed to complete the process. See also dispatching, priority-based scheduling, and thread.

sibling processes — The child processes of a single parent process. See also process and spawn.

signal — A data transmission event or group of events representing a bit or group of bits; a message sent from one active process to another.

signal-to-noise (S/N) ratio — A mathematical relationship between the power of a carrier signal and the power of noise in the communication channel; measured in decibels.

signal wire — In electrical transmission through wires, the channel component used to carry data.

signed integer — An integer that uses a sign bit to indicate whether the value is negative or positive.

Simple Mail Transfer Protocol (SMTP) — The earliest e-mail protocol; defines how text messages are forwarded and routed between Internet hosts.

Simple Object Access Protocol (SOAP) — An open standard, developed by the World Wide Web Consortium; has a simple programming interface and few infrastructure requirements.

simplex mode — A communication mode in which messages flow in only one direction.

sine wave — A waveform that varies continuously between positive and negative states.

single inline memory module (SIMM) — A small printed circuit board that incorporates multiple DIPs and has a row of electrical contacts on the edge; the entire package is designed to lock into a SIMM slot on a motherboard. See also dual inline packages (DIPs).

single-mode cable — A fiber-optic cable with fibers that are much thinner in diameter than multimode fibers and vary continuously in density from center to edge to eliminate light reflections.

singly linked list — A data structure in which each list element contains a pointer to the next list element.

skew — The timing difference between the arrivel of bits sent in a parallel transmission channel; skew increases with distance and transmission rate. See also parallel transmission.

socket — The combination of an IPaddress with a port number, such as 129.24.8.4:53; used to establish connections.

software as a service (SaaS) — A Web-based or cloud-based architectural approach in which users interact via a Web browser or other Web-enabled view layer with application software provided by a third party and installed on the provider�s hardware.

software developers — People who create application software for specific processing needs; can include many job titles, with each role contributing to a different part of the SDLC. See also systems development life cycle (SDLC).

software monitor — A program typically embedded in OS service routines that detects and reports processing activity or requests; can also generate statistics of service utilization or processing activity that can be displayed in real time or stored in a file for later analysis.

solid-state drive (SSD) — A storage device that mimics the behavior of a magnetic disk but uses flash RAM or other nonvolatile memory devices as the storage medium and read/write mechanism; expected to gradually replace magnetic disks.

sound card — An expansion card connected to the system bus that contains components for sound input and output.

source code — Instructions or statements in a high-level programming language; normally stored in a file that�s named to indicate both its function and programming language.

spawn — The act of a parent process creating a child process. See also process.

speaker dependent — Requiring training to recognize the sounds of human speakers; a characteristic of speech-recognition systems. See also speech recognition.

special-purpose processor — A processor that�s designed to perform only one specific task; essentially, a processor with a single internal program. See also processor.

special-purpose registers — Registers used by the CPU for specific tasks; they include the instruction register, instruction pointer, and program status word. See also registers.

speculative execution — An approach to dealing with condition BRANCHes in which the CPU executes instructions after a branch prediction but before the final branch condition value is known with certainty; a form of parallel processing. See also branch prediction.

speech recognition — The process of recognizing and responding to the meaning embedded in spoken words, phrases, or sentences.

speech synthesis — A complex process for generating human speech based on character or text input.

stack — A reserved area of primary storage accessed on a last-in, first-out (LIFO) basis; this mechanism enables a program suspended by an interrupt to resume execution in exactly the same state as before an interruption.

stack overflow — An error condition that occurs when attempting to push register values onto a stack that�s already at its maximum capacity. See also stack.

stack pointer — A special-purpose register that points to the next empty address in the stack and is incremented or decremented automatically each time the stack is pushed or popped. See also stack.

star topology — A network topology in which every node is connected directly to a shared hub, switch, or router. See also network topology.

start bitsBits added to the beginning of messages in asynchronous transmission to alert the receiver to synchronize its clock.

stateful firewall — A firewall that tracks the progress of complex client/server interactions. See also firewall.

static connection — A mapping between a local resource name and a remote resource that must be initialized before use.

static linking — A linking process in which library calls and other functions can�t be changed after they�re inserted into executable code. See also link editor.

static RAM (SRAM) — A type of RAM that�s implemented entirely with transistors; the basic storage unit is a flip-flop circuit.

storage allocation table — A data structure that records which allocation units are free and which belong to files. See also allocation unit.

storage area network (SAN) — A high-speed interconnection between general-purpose servers and a separate storage server; storage accesses are at the level of disk sectors in a logical address space.

storage bus — A subsidiary bus that connects secondary storage devices to the system bus; reduces the length and number of physical connections to the system bus and aggregates the lower data transfer capacity of secondary storage devices to better match the higher capacity of a single system bus connection.

storage I/O control layer — An FMS layer; the part of the kernel that accesses storage locations and manages data movement between storage devices and memory. Includes device drivers, interrupt handlers, and buffers and cache managers. See also file management system (FMS).

storage medium — A device or substance in a storage device that actually holds data.

store — A data transfer from a register into primary storage.

store and forward — An interconnected system of end nodes and transfer points used to route data between source and destination nodes.

strategic plan — A set of long-range goals for services to be provided and the resources needed to provide these services.

string — A sequence of characters forming a meaningful word, phrase, or other useful group.

subroutineSee function.

subtractive colors — In printing, the primary colors are generated by using the inverse of the primary video display colors, so cyan is the absence of red, magenta is the absence of green, and yellow is the absence of blue.

supercomputer — A computer system designed for rapid mathematical computations and used for computation-intensive applications, such as simulations, 3D modeling, weather prediction, computer animation, and real-time analysis of large databases.

Super Digital Linear Tape (SDLT) — A magnetic tape standard developed by Quantum Corporation; the cartridge has only one reel, and the device records in parallel linear tracks in an end-to-end format.

supervisor — The master interrupt handler program; it examines the interrupt code stored in the interrupt register, uses it as an index to the interrupt table, extracts the corresponding memory address, and transfers control to the interrupt handler at that address. See also interrupt code and interrupt handler.

surge protector — A hardware device that detects incoming power surges and diverts them to ground. See also power surges.

sustained data transfer rate — The maximum data transfer rate a storage device or communication channel can sustain during lengthy data transfer operations.

swap fileSee swap space.

swap space — A secondary storage region reserved for the task of holding pages not held in memory; it�s divided into page frames in the same manner as memory. See also page and virtual memory management.

switch — A central connection point for nodes in a LAN; examines incoming destination addresses and temporarily connects the sending transmission line to the receiving transmission line. Also refers to a building block of processing circuitry; it controls electrical current flow in a circuit and is implemented as a transistor.

switched fabric — An architecture for interconnecting devices with multiple data transmission pathways and a mesh of switches resembling the interwoven threads of fabric.

symbol table — An internal table updated by the compiler that keeps track of data names, types, and assigned memory addresses in programs. See also compiler.

symbolic debugger — An automated tool for testing executable programs; includes features for tracing calls to source code statements or functions, tracing changes to variables� contents, and detecting runtime errors.

synchronous DRAM (SDRAM) — A read-ahead RAM that uses the same clock pulse as the system bus; read and write operations are broken into simple steps that can be completed in one bus clock cycle.

synchronous idle characters — Control messages consisting of predetermined pattern of signal transitions designed for easy clock synchronization.

synchronous transmission — A method that ensures sender and receiver clocks are always synchronized by sending data in continuous streams of fixed-size byte groups called �blocks.�

system administration — A wide range of managerial activities for ensuring efficient and reliable delivery of information services.

system bus — The internal communication channel connecting all hardware devices.

system clock — A digital circuit that generates timing pulses (signals) and transmits them to other devices in the computer; all actions, especially a CPU�s fetch and execution cycles, are timed according to this clock.

system development tools — Tools that enable systems analysts and designers to develop models of information systems that are then used as the starting point for developing application programs. See also application development software.

system overhead — The resources consumed by resource allocation procedures.

system requirements models — Models that provide the detail needed to develop a system that meets users� needs.

system software — A program for handling resource allocation to application software, performing utility functions needed by application software, or managing computer resources; includes operating systems, database management systems, antivirus software, and network security software, for example.

systems analyst — A software developer who contributes to the business modeling and requirements disciplines. See also software developers.

systems architecture — The structure, interaction, and technology of computer components.

systems designer — A software developer who contributes to the design discipline. See also software developers.

systems development life cycle (SDLC) — The process for developing an information system; follows a series of steps or activities. See also Unified Process (UP).

systems programmer — Person who develops or maintains system software; also performs tasks such as hardware troubleshooting and software installation and configuration. See also system software.

tape drive — A slow serial access device containing motors that wind and unwind tapes and read/write heads to access tape content.

target channel adapter (TCA) — An interface that connects a device to an InfiniBand switch; used by simpler devices, such as network switches and storage appliances.

TCP/IPSee Internet Protocol (IP) and Transmission Control Protocol (TCP).

TCP/IP model — A layered protocol model describing current Internet standards and technology. See also Internet Protocol (IP) and Transmission Control Protocol (TCP).

Telnet — A Web protocol in which users on one Internet host can interact with another host�s OS command layer; it emulates a VDT and is limited to interacting with command-line interfaces.

testing discipline — Activities in the Unified Process for verifying the correct functioning of infrastructure and application software components and ensuring that they satisfy system requirements. See also Unified Process (UP).

thin film transistor (TFT) — A technology for manufacturing active matrix displays, in which wiring and transistors are added in thin layers to a glass substrate; similar to semiconductor fabrication technology.

thread — A subdivision of a process that can be scheduled and executed independently; shares all resources allocated to its parent process. See also process.

thread control block (TCB) — A data structure containing information that an OS uses to keep track of and manage threads. See also thread.

thread cycle — The amount of CPU time needed to complete a thread�s function. See also thread.

thread listSee run queue.

three-layer architecture — A variation of client/server architecture that divides software into three client or server processes called layers: the view layer, the business logic layer, and the data layer.

three-tier architectureSee three-layer architecture.

third-generation language (3GL) — A high-level programming language that uses mnemonics to represent instructions, variables, and labels; is machine independent; and has instruction explosion higher than 1:1. It has no advanced capabilities for interactive I/O, database processing, or nonprocedural programming. See also instruction explosion.

time-division multiplexing (TDM) — A technique that divides a channel�s data transfer capacity into time slices and allocates them to multiple users.

timer interrupt — An interrupt generated at regular intervals (between several dozen and several thousand CPU cycles) to give the scheduler an opportunity to suspend the currently executing thread. See also scheduling and thread.

traces — Arrangements of conductive molecules (usually straight lines) that enable electrons to flow from one place or device to another.

track — One concentric circle of a platter; the surface area that passes under a read/write head when its position is fixed. See also platters.

track-to-track (TTT) seek time — The average time needed to move a read/write head between two adjacent tracks; typically measured in milliseconds.

transaction — To an FMS, any change to file contents or attributes, such as an added record, a modified field, or changed access controls. See also file management system (FMS).

transaction logging — A form of automated backup in which all changes to file content and attributes are recorded automatically in a log file in addition to being written to the file�s I/O buffer; provides a high degree of protection against data loss caused by program or hardware failure imposes a performance penalty.

transistors — Electrical switches made of semiconductor material that has been treated with chemical impurities to enhance the semiconducting effects; they�re combined to implement gates. See also gate.

Transmission Control Protocol (TCP) — A core Internet protocol for translating messages into packets and guaranteeing their delivery.

transmission medium — A communication path that carries signals.

Transport layer — The OSI layer responsible for formatting messages into packets suitable for transmission over the network; adds required header and trailer information, including network addresses, error-detection data, and packet-sequencing data. See also Open Systems Interconnection (OSI) model.

tree directory structureSee hierarchical directory structure.

trillions of floating-point operations per second (teraflops or TFLOPS) — A measurement of the rate at which floating-point operations are performed; used to measure CPU performance.

truncation — The act of deleting bits that don�t fit in a storage location.

twin-axial cable — A transmission medium that�s similar to coaxial cable but is thinner and contains two internal conductors. See also coaxial cable.

twisted-pair cable — A transmission medium consisting of two copper wires twisted around one another and encased in nonconductivematerial, usually plastic.

twos complement notation — A notation system that represents positive integers as ordinary binary values and negative integers by adding 1 to the complement of the positive value.

Type I error — In data transmission, the probability of not detecting a real error.

Type II error — In data transmission, the probability of incorrectly identifying good data as an error.

unblocked — The term describing a physical record containing just one logical record. See also logical record and physical record.

unconditional BRANCH — An instruction in which the processor always departs from the normal execution sequence. See also BRANCH.

undelete operation — The act of restoring a record or file by re-creating its index information (directory entry) and recovering its previously allocated storage locations.

underflow — A condition that occurs when a value is too small to represent in floating-point notation; also refers to overflow of a negative exponent in floating-point notation. See also floating-point notation.

unguided transmission — A transmission medium that uses the atmosphere or space to carry messages encoded in radio frequency or light signals; also called wireless transmission.

Unicode — A standard 16-bit or 32-bit character-coding method that assigns nonnegative integers to represent printable characters; includes alphabets, ideographs, and characters for most of the world�s languages.

Unified Process (UP) — A systems development life cycle based on object-oriented techniques; follows a series of repeated steps. See also iterations.

Uniform Resource Locator (URL) — A unique identification for a Web resource, composed of a protocol, host, port, and resource.

uninterruptible power supply (UPS) — A device that provides power to attached devices in the event of external power failure; UPSs vary in their power delivery capacity, switching time, and battery life.

unresolved referenceSee external function call.

unsigned integer A data type that stores positive integer values as ordinary binary numbers; its value is always assumed to be positive.

User Datagram Protocol (UDP) — A connectionless protocol that lacks the connection management features of TCP but can support communication between multiple hosts and senders; used mainly for communication that requires low processing overhead and doesn�t require a guarantee of reliable delivery. See also connectionless protocol.

variable — A mnemonic representing a data item�s memory address in assembly language or a name representing a data item�s memory address in a high-level programming language. See also assembly language.

variable-length instruction — In this type of instruction format, the amount by which the instruction pointer is incremented after a fetch is the length of the most recently fetched instruction; this format complicates instruction fetching because the number of bytes to be fetched isn�t known in advance.

vector — In graphics, a line segment with a specific angle and length in relation to a point of origin; also refers to a one-dimensional array.

vector list — A series of concatenated or linked vectors that can be used to construct complex shapes; images constructed from a vector list resemble connect-the-dots drawings.

versioning — A process in which a file�s original version is archived automatically whenever the file is modified.

vertical redundancy checking See parity checking.

victim — If all page frames are allocated, the page currently in memory that must be written to the swap space before the reference page is loaded into a page frame. See also page, swap space, and virtual memory management.

video bus — A subsidiary bus that connects only memory and the video display device; improves performance by removing display update traffic from the system bus and providing a high-capacity one-way communication channel optimized for video data.

video controller — A device connected to the system bus (or a dedicated video bus) that accepts commands and data from the CPU and generates analog or digital video signals, which are transmitted to the monitor.

video display terminal (VDT) — An early I/O device containing an integrated keyboard and TV screen; capable of displaying only text and primitive graphics.

video RAM (VRAM) — A type of RAM type used in a video controller; differs from ordinary RAM because it can be written by the bus interface circuitry or video processor while being read by display generator circuitry. See also dual-porting.

view layer — The software layer that accepts user input and formats and displays processing results. See also three-layer architecture.

virtualization environment — Hypervisor that is installed as an application within another operating system. See also hypervisor.

virtual machine (VM) — A collection of files on a physical computer that define the virtual machine�s configuration and the contents of its virtual disk drives; creates an environment separate from the physical computer in which different OSs can run, application software can be tested, and so forth.

virtual memory management — A memory management method in which the OS divides memory and programs into partitions called pages, which are held in secondary storage until needed. See also page.

virtual resources — Resources (hardware and system software) that are apparent to a user or program as being available but don�t necessarily exist physically.

virus — A program or program fragment that infects a computer by installing itself permanently, performs malicious acts on the infected computer, and replicates and spreads itself by using services on the infected computer.

Voice over IP (VoIP) — A family of technologies and standards for carrying voice messages and data over a single packet-switched network.

volatile — A term describing storage devices that can�t hold data for long periods; primary storage is usually volatile.

volume — Part of the logical view of secondary storage; consists of an entire physical disk, a partition of the disk, or a removable storage medium and on larger computers, can span multiple physical disks.

wait state — An idle clock cycle during which the CPU is waiting for a response from another device. See also clock cycle.

wavelength-division multiplexing (WDM) — In long-distance telecommunication, using FDM to multiplex single-mode optical fibers. See also frequency division multiplexing (FDM).

wide area network (WAN) — A network that spans large physical distances, such as multiple buildings, cities, regions, or continents.

wired transmissionSee guided transmission.

wiresSee traces.

wireless transmissionSee unguided transmission.

word — A unit of data containing a fixed number of bytes or bits, loosely defined as the amount of data a CPU processes at one time; word size normally matches the size of general-purpose registers. See also general-purpose registers.

working directorySee current directory.

workstation — A type of microcomputer that�s typically more powerful than a PC to support demanding numeric or graphics processing tasks. See also microcomputer.

Worldwide Interoperability for Microwave Access (WiMAX) — A group of wireless networking standards codified in the IEEE 802.16 standard; targeted to applications involving fixed and mobile Internet access spanning distances up to 50 kilometers (about 30 miles). See also metropolitan area networks (MANs).

World Wide Web (WWW) — Also referred to as just �the Web,� a collection of resources that can be accessed over the Internet by standard protocols, such as HTTP.

This page was last modified on October 06, 2014.

Click here to return to the top level page of this web site.