Network Terms

By Dave Harrold, CONTROL ENGINEERING May 1, 1999

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

32-bit – Refers to number of bits used by an operating system to perform an operation. Is based on the microprocessor that the OS is designed to run on.

access method – Technique for moving data between main storage and input/output devices. In a Systems Network Architecture (SNA) environment, it is the software that controls the flow of information in a network.

adapter – Hardware installed in a PC or other computer and used to connect the computer to other hardware.

address – Identifier assigned to networks, stations and other devices so that each device can be separately designated to receive and reply to messages.

Address Resolution Protocol (ARP) – Internet protocol that dynamically maps Internet addresses to physical (hardware) addresses on local area networks. ARP is limited to networks that support hardware broadcast.

Advanced Program-to-Program Communications (APPC) – Part of the SNA protocol that establishes the conditions that enable programs to communicate across the network. This capability, involving LU6.2 and its associated protocols, allows communication between two or more processes in an SNA network without the involvement of a common host system or of terminal emulation.

agent – The part of a networked system that performs information preparation and exchange on behalf of a software entity.

algorithm – A prescribed set of well-defined rules or processes for arriving at a solution to a problem.

American National Standards Institute (ANSI) – ANSI is responsible for the establishment of many standards, including a number of data communications and terminal standards. ANSI is the recognized U.S. representative within CCITT and ISO. See also CCITT and ISO.

American Standard Code for Information Interchange (ASCII) – A 7-bit code, intended as a U.S. standard for the interchange of information among communications devices.

AppleShare – AppleShare is Apple Computer’s networking solution. It requires a dedicated Macintosh as a network server and includes both server and workstation software. It uses the AppleTalk Filing Protocol (AFP).

AppleTalk – AppleTalk is a set of communications protocols (such as IPX/SPX and NCP) used to define networking on an AppleShare network. On the OSI model, AppleTalk is comparable to communications protocols, in that both protocols specify communications, ranging from application interfaces to media access.

AppleTalk Filing Protocol (AFP) – Allows distributed file sharing across an AppleTalk network.

application – A software program or program package that makes calls to the operating system and manipulates data files, thus allowing a user to perform a specific job (such as accounting or word processing). application binary interface (ABI) A specification defining the interface between an operating system and a certain hardware platform, particularly the calls between applications and the operating system.

application interface – A set of software routines and associated conventions that permits application programmers to use that interface as a part of any application. In general, an application interface is used to access system or networking services that would otherwise require significant development effort to create from scratch. For example, the ManageWise application interface lets a programmer use ManageWise file structures and services within an application. See also application programming interface.

application programming interface (API) – A means by which an application gains access to system resources, usually for the purpose of communication (the sending and receiving of data), data retrieval or other system services. In the specific area of terminal emulation, an API provides for the simulation of keystrokes and for writing into and reading from the presentation space (device buffer). It may also provide for the sending and receiving of structured fields.

application server – A server in a client-server network which runs one or more applications that can be shared by client stations and which also shares the data processing burden with client stations. This shared application and shared data processing model contrasts with the model used for other servers, such as file servers, that simply send, receive, and store files, requiring client stations to run all applications and process all data. Either model can be most advantageous, depending on circumstances. In many circumstances the application server model allows for faster data processing, faster throughput to client stations, greater data reliability, and increased data security.

architecture – The manner in which a system, such as a network, computer or program is structured.

archive – To create a redundant copy of computer file data, typically to create a backup copy of that data to protect it if the original copy is damaged or otherwise irretrievable. By some definitions, an archive is required to contain copies of every version of a particular file. In this case, to archive means to save a copy of every object in a file system with a separate copy of all changes made to that file. In addition to protecting files from loss, this approach also permits any previous version of a file to be restored, typically by date and time.

ARCnet (Attached Resource Computing Network) – A proprietary token-bus networking architecture developed by Datapoint Corporation in the mid-1970s. Currently, ARCnet is widely licensed by third-party vendors and is a popular networking architecture, especially in smaller installations. It is relatively fast (2.5 Mbit/s) and reliable, and it supports coaxial, twisted pair and fiber optic cable-based implementations.

ASCII – See American Standard Code for Information Interchange.

asynchronous – A data transmission method in which each character is sent one bit at a time. Each character has a start and stop bit to synchronize signals between the sending device and receiving device. This allows a character to be sent at random after the preceding character has been sent. See also synchronous.

autoauthentication – In a client-server environment, a utility that lets users access unrestricted network resources without password verification. Only when a user attempts to access a restricted resource does the utility prompt for a password.

autologin – In a network environment, a utility that regulates user login attempts.

automount – A graphical utility that provides an iconical tree structure to simplify the user’s task of locating and using a server, filesystem, or volume.

backbone network – Primary connectivity mechanism of a hierarchical, distributed network. Ensures that all systems that have connectivity to an intermediate system on the backbone have connectivity to one another.

bad block table – A list kept on a hard disk of storage locations on the disk that are damaged or physically unable to hold data reliably. The bad block table is usually duplicated on a label on the outside of the disk housing. Also called ‘media defect list.’ See also Hot Fix, read-after-write verification.

bandwidth – Carrying capacity of a circuit, usually measured in bits per second for digital circuits, or hertz for analog circuits.

baseband – Network technology that uses a single carrier frequency and requires that all stations attached to a network participate in every transmission.

basic input/output system (BIOS) – A set of programs, usually in firmware, that enables each computer’s central processing unit to communicate with printers, disks, keyboards, consoles and other attached input and output devices.

beaconing – In token-ring networks, the state that results when an error condition occurs, preventing communication until the error condition is resolved.

Berkeley Software Distribution (BSD) – A UNIX operating system version developed at the University of California, Berkeley.

binary – Numbering system using only zeros and ones.

bindery – A network database, in versions of the network operating system earlier than 4.0, that contains definitions for entities such as users, groups, and workgroups.

BIOS – See basic input/output system. bit Binary digit; either a one or a zero.

bit/s – The rate at which data is transferred over a serial interface.

block – Set of continuous bits or bytes that make up a definable quantity of information such as a message.

bridge – See router.

broadcast – Packet delivery service in which all nodes on a network receive a copy of any frame that is designated for broadcast or, when used as a verb, sending the message to all nodes.

broadband – Characteristic of any network that multiplexes multiple, independent network carriers on a single cable. Allows several networks to coexist on a single cable. Traffic from one network does not interfere with traffic from another network because conversations happen on different frequencies.

buffe r – Memory area or electronic register where data is stored temporarily while awaiting disposition. It compensates for differences in data-flow rates (for example, between a terminal and its transmission line). Also used as a data backup mechanism, holding data that may be retransmitted if an error is detected during transmission.

burst – Method of data transfer in which information is collected and sent as a large unit in one high-speed transmission. LAN traffic is usually considered bursty traffic because it has short intervals of intense activity with lulls between. byte Group of eight binary digits operated on as a unit; also known as a character or octet.

cache – High-speed memory section that holds blocks of data that the CPU is currently working on; designed to minimize the time the CPU spends accessing memory.

central processing unit (CPU) – Main processing unit of a computer.

CCITT recommendations – The CCITT is a committee that recommends standards for communications equipment interfaces, communications protocols, modem modulation methods and so on.

channel – Path for transmitting electromagnetic signals; synonym for line or link.

channel service unit (CSU) – Digital signal processor that performs transmit and receive filtering, signal shaping, longitudinal balance, voltage isolation, equalization, and remote loopback testing for digital transmission. It functions as a guaranteed safe electrical circuit, acting as a buffer between the customer’s equipment and a public carrier’s wide area network. CSUs prevent malfunctioning digital service units (DSUs) or other customer premises equipment from disabling a public carrier’s transmission system. The design of a CSU must be certified by the FCC.

character – Group of eight binary digits operated on as a unit; also called a byte or octet.

circuit – Any path that can carry an electrical current.

client – 1.) Node or workstation (computer) on a computer network that requests services from a network server. Client stations run client software. 2). An executing software program or set of programs through which a client station sends a request to a server and waits for a response (for example, the Novell NetWare Client for DOS/MS Windows). 3.) The user end of a client-server connection.

client-server model – 1.) Data communication model that relies on distributed, intelligent interaction between network servers and individual (client) workstations. Clients request services from servers. Servers receive client requests and return requested data or results. Clients and servers may be any class of computer, but often the client is a desktop computer and servers are powerful microcomputers, ‘workstation class’ computers or minicomputers. 2.) Data communication model in which there is (1) server software that starts execution before communication begins and, after communication begins, continually accepts requests from and returns responses to clients, and (2) client software that periodically sends requests for services to the server and accepts server responses.

client-server network – A network consisting of client nodes (workstations) which have client capabilities only and server nodes which have (usually) server capabilities only. On a client-server network, communication and data sharing between clients is, in most cases, arbitrated by the network servers. Each client runs client software and each server runs a client-server operating system (see client-server operating system below). A simple network with only one server, a file server, is a classic example of a client-server network.

client-server operating system – An operating system which runs on a server in a client-server network and which is responsible for coordinating the use (by clients) of all resources available from that server. Server resources include hardware such as hard disks, Random Access Memory (RAM), printers and equipment used for remote communications, such as modems. Resources also include logical systems such as the network file system and network directory services, and the information (data) therein.

cluster controller – An IBM or IBM-compatible device for the attachment of 3270 or 3270-class terminals. May be channel-attached to a host system or may communicate with the host via an SDLC or, in some cases, a bisynch link to a host-attached communications controller.

coaxial – A type of cable that uses two conductors: a central, solid wire core, surrounded by insulation, which is then surrounded by a braided wire conductor sheath. This cable is particularly well suited for networking because it can accommodate high bandwidth but is relatively resistant to interference.

collision – What happens when two devices transmit data at the same time, resulting in a loss of data.

compression – Method of compacting data into a smaller number of bits for more efficient transmission or storage.

concentrator – Device with a single bus and multiple connections to computers; provides a star-wired physical layout.

configuration – The way in which a system or part of a system, such as a piece of software, is set up, based on a number of possible choices.

configuration management – Refers to the management of networked applications and their related user access.

connectionless – Model of interconnection in which communication takes place without first establishing a connection.

Connectionless Network Protocol (CLNP) – Open Systems Interconnection (OSI) protocol that provides the OSI Connectionless Network Service (delivery of data).

Connectionless Transport Protocol (CLTP) – Provides end-to-end transport data addressing and error detection, but does not guarantee delivery or provide flow control. The Open Systems Interconnection (OSI) equivalent of the User Datagram Protocol (UDP) datagram service.

connection-oriented – Model of interconnection in which communication proceeds through three well-defined phases: connection establishment, data transfer, and connection release.

connectivity – The ability to connect to and communicate with multiple architectures on a single network.

console – The monitor and keyboard from which you actually view and control server or host activity.

controller board – See adapter.

data – Data are entities that convey meaning. Computer data is stored as a series of (electrical) charges arranged in patterns to represent information. In other words, data refers to the form of the information (the electrical patterns). It is not the information itself.

database – Set of logically connected files that have a common access. All data entities that exist for several related systems. A database can have several data items that can be assembled into many different record types.

data compression devices – Equipment that compresses data into a smaller number of bits, allowing more data to be transmitted per second than the link speed otherwise allows.

Data Encryption Standard (DES) – A standard encryption technique that scrambles data into a code for transmission over a public network.

datagram – One packet of information and associated delivery information that is routed through a packet-switching network.

data rate – Speed at which data bits are transmitted and received. Usually measured in bits per second.

DECnet – A set of networking protocols developed by Digital Equipment Corporation and used in its VAX family of computers to exchange messages and other data. Although DECnet is currently a proprietary protocol, DEC is merging its protocols with OSI protocols for the upcoming DECnet Phase V. When this process is complete, DECnet protocols should interoperate with any OSI-compliant network node.

decryption – Unscrambling or decoding of encrypted data.

dedicated – A device that has only one function. For example, a dedicated server cannot be used as a workstation. See also nondedicated.

dedicated line – Leased or private communications line. See also dial-up line.

demodulation – Process of recovering information from a previously modulated carrier frequency by converting analog signals into digital signals.

desktop computer – A small-scale computer that fits on a desktop and that has a microprocessor system. Also called a microcomputer or personal computer. Contrast with minicomputer and mainframe computer.

developer – One who develops software, either for internal use or for commercial sale.

device driver – Software or firmware that translates operating system requests (such as input/output requests) into a format that is recognizable by specific hardware, such as adapters.

dial-up line – Communications line accessible via dial-up facilities, typically the public telephone network. See also dedicated line.

digital – Representation of information using ones and zeros. It is discretely variable as opposed to continuously variable. Data characters are coded in discrete separate, electrical pulses or signal levels. Contrast with analog.

digital service unit (DSU) – Device between a user’s data terminal equipment (DTE) and a common carrier’s digital circuits. It formats data for transmission on public carrier wide area networks and ensures that the carrier’s requirements for data formats are met.

directory caching – Feature to improve performance. In directory caching, copies of the file allocation table and the directory entry table are written into the network server’s memory. A file’s location can then be read from memory, which is faster than reading it from a disk.

directory hashing – Feature to improve performance. Directory hashing is a method of indexing file locations on a disk so the time needed to locate a file is reduced.

directory rights – Restrictions specific to a particular directory.

directory services – Network service that provides information about an entity of interest. It is like an electronic phone book to help network clients find services. There are several designs, including the X.500 standard, the Domain Name System and Novell’s NetWare Directory Services.

directory structure duplication – Feature that protects data from failures in network hardware. A hard disk’s directory and file allocation tables contain the address information the operating system needs to determine where to store or retrieve data. To reduce the possibility of losing this information, the network operating system (NOS) maintains duplicate copies of both the directory table and the file allocation table on separate areas of the hard disk. If the primary copy is lost or destroyed, the NOS uses the secondary copy.

directory verification – Feature that protects data from failures in network hardware. Each time the server is turned on, the network operating system performs a consistency check on duplicate sets of directory and file allocation tables to verify that the two copies are identical.

disk duplexing – Feature that protects data from failures in network hardware. In disk duplexing, all data on one hard disk is duplicated on a second hard disk on a separate channel. Disk writes made to the original disk are also made to the second disk. If the original disk or channel fails, the duplicate disk takes over automatically.

disk mirroring – Feature that protects data from failures in network hardware. In disk mirroring, all data on one hard disk is duplicated on a second hard disk on the same channel. Disk writes to the original hard disk are also written to the second hard disk. If the original disk fails, the duplicate disk takes over automatically.

distributed application – An application that operates in a distributed computing environment, where application modules may run on different systems.

distributed computing – A computer operating environment that may involve computers of differing architectures and data representation formats that share data and system resources.

distributed network – A computer network on which processing is shared by many different parts of the network. Processing may be shared by client (local) computers, file servers, print servers and application servers such as database servers. Distributed processing enables the most efficient use of processing power because available processors can be dynamically assigned as either general or job specific processors, depending on the type of work to be done and the existing work load. Distributed processing also enables duplication and distribution of key services, such as directory services, so that full services remain available regardless of the failure of individual parts of the network.

distributed processing – A technique to enable multiple computers to cooperate in the completion of tasks, typically in a networked environment. Each computer that contributes to the completion of the total task actually does so by completing one or more individual subtasks independently of its peers, reporting the results from its subtasks as they are completed.

domain – In the Internet, a part of a naming hierarchy. Syntactically, an Internet domain name consists of a sequence of names separated by periods. In the network operating system and OSI, it is generally used as an administrative partition of a complex distributed system.

Domain Name System (DNS) – Distributed name/address database used on the Internet.

DOS – A generic term to refer to those operating systems that use commands rather than having a graphical user interface. The most common of these are DR DOS, MS-DOS and PC DOS.

downsizing – The trend to off-load some applications from proprietary mainframe systems to smaller, less expensive, networked microcomputers.

downtime – Time when a system or network is unavailable.

drag-and-drop function – A mouse pointer operation in which you select an object and place it in a new location. For example, to print a document, you drag its icon to the printer icon and drop it there.

driver – See device driver.

dumb terminal – Simple CRT and keyboard with limited capabilities such as display and edit functions.

elevator seeking – Feature to improve performance. Elevator seeking allows the disk read-write head to pick up files in the direction it is traveling across the disk rather than picking them up in the order they were requested.

E-mail (electronic mail) – A method of file transfer and message sending among workstations. encryption Scrambling or coding of data for security.

engine – The core of a database or of an application.

error detection – Process of determining whether one or more bits have changed from a one to a zero, or vice versa, during transmission.

Ethernet – A network cable and access protocol scheme originally developed by DEC, Intel and Xerox but now marketed primarily by DEC and 3Com.

EtherTalk – AppleTalk packets encapsulated to run on Ethernet cables.

expansion – Increasing the capability of a microcomputer by adding hardware that performs a task that cannot be done with the basic system.

Extended Binary Coded Decimal Interchange Code (EBCDIC) Eight-bit code defined by IBM. Includes values for control functions and graphics.

FCONSOLE – Utility used to access information from the network server and fine-tune its performance. It is a virtual console utility that allows the operator to control a server from any station on the network.

fiber-optic cable – High-bandwidth transmission medium that allows data to be transmitted by modulating a light wave through a special glass or plastic fiber.

file allocation table (FAT) – A FAT keeps track of file locations in a particular volume. The network operating system (NOS) divides each volume into blocks and stores files on the volume in these blocks. If the file consists of one or more blocks, the file may be stored in blocks that are not adjacent. The FAT keeps track of the block numbers where different parts of the file are located. To retrieve a file, the NOS searches through the FAT until it finds the FAT entries and corresponding block numbers for the requested file.

file sharing – An important feature of networking that allows more than one user to access the same file at the same time.

File Transfer, Access, and Management (FTAM) – The Open System Interconnection (OSI) remote file service and protocol.

firmware Set of software instructions that are set, permanently or semipermanently, into integrated circuitry.

gateway – A hardware/software package that runs on the OSI application layer and allows incompatible protocols to communicate; includes X.25 gateways. Usually connects PCs to a host machine, such as an IBM mainframe.

gigabyte (GB) – A unit of measure for memory or disk storage capacity. Ten to the ninth power (one billion) bytes.

groupware – A type of software that supports concurrent use of objects (such as documents, calendars and spreadsheets) by multiple LAN users.

Hierarchical File System (HFS) – Attached to AFP in the Macintosh operating system. It manages files and directories.

High-Level Data Link Control (HDLC) – Communications protocol defined for high-level, synchronous connections to X.25 packet networks. Similar in almost all respects to SDLC. See also synchronous.

High-Level Language/Application Program Interface (HLLAPI) – Application programming interface designed for use with high-level languages.

host – A computer, attached to a network, that provides services to another computer beyond simply storing and forwarding information. Usually refers to mainframe and minicomputers.

Hot Fix – Feature that protects data from failures in network hardware. When the Hot Fix feature is activated, a small portion of a hard disk’s storage space is set aside as a Hot Fix redirection area. When read-after-write verification determines that there is a bad data block on the disk, Hot Fix redirects data that was to be stored in the bad block to the Hot Fix redirection area. Hot Fix marks the defective block as bad, and the server will not attempt to store data there again.

hub – Concentrator or repeater at which node connections meet in a star physical layout.

hypertext – A method for storing, retrieving and presenting information based on the processing power of computers. Allows computerized linking and almost instantaneous retrieval of information based on a dynamic index.

Institute of Electrical and Electronic Engineers (IEEE) – Creates networking standards for cabling, electrical topology, physical topology and access schemes.

intelligent hub – Unit combining the function of a hub with processing capabilities.

Integrated Services Digital Network (ISDN) – A CCITT standard that covers a wide range of data communications issues but primarily the total integration of voice and data. Already having major effects on exchange and multiplexer design.

interface – Point at which a connection is made between two elements so that they can work together.

International Standards Organization (ISO) – Based in Paris, this organization develops standards for international and national data communications.

Internet – Collection of networks and gateways that use the TCP/IP suite of protocols. Lowercase, it is an abbreviation for internetwork.

internetwork – Two or more networks connected by an internal or external router.

Internetwork Packet Exchange (IPX) – A protocol that allows the exchange of message packets on an internetwork.

interoperability – Ability for devices on a heterogeneous network to transmit and share data.

jabber – State of a network adapter in which a network device continuously transmits.

journaling – The process of logging system activity to facilitate a fast restart when needed.

kernel – The core of an operating system that is responsible for managing system resources.

kilobits per second (kbit/s) – Unit of measure for data transfer rates; two to the 10th power (1,024) bits per second.

kilobyte (KB) – A unit of measure for memory or disk storage capacity; two to the 10th power (1,024) bytes.

kilobytes per second (kbyte/s) – One thousand twenty-four bytes per second. Unit of measure commonly used for transfer rates to and from peripheral devices.

local area network (LAN) – A system that links computers together to form a network, usually with a wiring-based cabling scheme. LANs connect personal computers and electronic office equipment, enabling users to communicate, share resources such as data storage and printers, and access remote hosts or other networks.

LocalTalk – Shielded twisted-pair cable introduced by Apple.

logical unit (LU) – Terminal emulation program or application in an SNA network. LUs can communicate with host systems and applications (LU Type 0, 1, 2, 3, 4 and 7) or with other LUs of the same type (LU Type 6.0, 6.1 and 6.2 only).

login script – A set of instructions that directs your workstation to perform specific actions when you log in to the network. The network supervisor can create a system-wide login script (which is the same for all users on the network) that instructs all workstations to perform the same actions upon login. Your individual login script executes after the system-wide login script. It specifies your individual drive mappings.

mainframe computer – A large-scale computer (such as those made by Burroughs, Control Data, IBM, Univac and others) normally supplied complete with peripherals and software. Also called a host or CPU. Contrast with minicomputer and desktop computer.

media – Plural of medium. Physical paths over which communications flow, such as copper wires, coaxial cable or optical fiber.

megabits per second (Mbit/s) – Unit of measure for data transfer rates; two to the 20th power (1,048,576) bits per second.

megabyte (MB) – A unit of measure for memory or disk storage capacity; two to the 20th power (1,048,576) bytes.

message – Logical grouping of information at the application layer.

minicomputer – A small-scale or medium-scale computer (such as those made by Data General, DEC, Hewlett-Packard and others) that usually services dumb terminals. Contrast with mainframe computer and desktop computer.

modem – Literally modulator/demodulator. Converts digital data into analog (waveform) signals for transmission along media that carry analog signals and converts received analog signals back into digital data for use by the computer. With the advent of digital lines, there is a also a new kind of modem, called a digital modem, that doesn’t actually modulate or demodulate signals but is merely responsible for their transmission over digital lines.

multicast – Special form of broadcast in which copies of the packet are delivered to multiple stations, but only a subset of all possible destinations.

multiple name space support – The method that allows various workstations running different operating systems to create their own familiar naming conventions. Different operating systems have different conventions for naming files, but with multiple name space support, the name spaces supported on a volume are configurable so that each file on a given volume has a name that any workstation can recognize.

Multiple Virtual Storage (MVS) – IBM operating system for large host systems.

multiplexer – Device that allows a single communications circuit to take the place of several parallel ones; often used to allow remote terminals to communicate with front-end processor ports over a single circuit.

multitasking – The ability to run two or more programs (tasks) on one computer at the same time. The tasks take turns using available I/O and CPU cycles.

multivendor network – Network comprised of components from different vendors.

NetBIOS (Network Basic Input/Output System) – A programmable entry into the network that allows systems to communicate over network hardware using a generic networking API that can run over multiple transports or media.

NetView – IBM network monitoring software for SNA networks.

network – A system that sends and receives data and messages, typically over a cable. A network enables a group of computers to communicate with each other, share peripherals (such as hard disks and printers), and access remote hosts or other networks.

network adapter – The hardware installed in workstations and servers that enables them to communicate on a network. See also adapter.

network computing – A multivendor computing environment that integrates local and wide area network technologies to provide enterprise-wide connectivity.

Network File System (NFS) – A distributed file system network protocol developed by Sun Microsystems.

node – Device that is connected to a network and is capable of communicating with other network devices.

noise – Unwanted changes in waveform that occur between two points in a transmission circuit.

nondedicated – A device that performs multiple simultaneous functions. For example, a nondedicated network server runs the network functions and performs as a workstation. See also dedicated.

open architecture – An architecture that is compatible with hardware and software from any of many vendors.

OPEN LOOK – AT&T’s non-proprietary GUI standard for 2-D and 3-D graphics, implemented by Sun Microsystems; includes a window manager and a toolkit.

Operation, Administration and Management (OA&M) package – A UnixWare menu-based interface to a suite of system administration and maintenance utilities.

operating system (OS) – Software that manages a computer system. It controls data storage, input and output to and from the keyboard and other peripheral devices, and the execution of compatible applications.

OS/2 – An operating system that uses a graphical user interface and was designed by IBM.

OSF/Motif – The Open Software Foundation’s non-proprietary GUI standard; includes a window manager and a toolkit.

OSI (Open Systems Interconnection) reference model – A model for network communications consisting of seven layers that describe what happens when computers communicate with one another.

packet – The unit of information by which the network communicates. Each packet contains the identities of the sending and receiving stations, error-control information, a request for services, information on how to handle the request and any necessary data that must be transferred.

packet assembler/ disassembler (PAD) – Device or program used to create packets of data for transmission over a CCITT X.25 packet data network and to remove data from the received packets. The most common is a CCITT X.29 PAD, used for packetizing and depacketizing asynchronous ASCII data.

peer-to-peer communication – Communication directly between devices that operate on the same communications level on a network, without the intervention of any intermediary devices such as a host or server. A peer-to-peer communication method (or protocol) defines only the basic mechanisms used to transfer data; it need not specify when or why peer application programs or nodes interact or how communication between such applications or nodes should be organized in a distributed environment. The latter problems fall into the domain of the peer-to-peer operating system (see peer-to-peer network below).

peer-to-peer network – A network consisting of nodes (computers) which all have both client and server capabilities and on which communication and data sharing is carried on directly between nodes, rather than being arbitrated by an intermediary node. On a peer-to-peer network all nodes run the same peer-to-peer operating system, which gives them both client and server capabilities.

performance tuning – Monitoring and analyzing the net performance of a system and adjusting its configuration to obtain optimum performance.

physical address – Data-link layer address of a network device.

physical unit (PU) – A node in an SNA network supporting one or more logical units (LUs).

pkgadd – A UnixWare command that simplifies the process of installingoperating system extensions and third-party software.

platform – Term used as a generic reference to all possible choices for some specific part of the computing environment. For example: desktop operating system platform (could include DOS, OS/2, and so on) or network operating system platform (NetWare, LAN Manager, and so on).

port – For hardware, a connecting component that allows a microprocessor to communicate with a compatible peripheral. For software, a memory address that identifies the physical circuit used to transfer information between a microprocessor and a peripheral.

protocol – Set of rules that allow computers to connect with one another, specifying the format, timing, sequencing and error checking for data transmission. protocol suite – A collection of networking protocols that provides the communications and services needed to enable computers to exchange messages and other information, typically by managing physical connections, communications services and application support.

query – Process of extracting data from a database and presenting it for use.

queue – A line or list formed by items waiting for service, such as tasks waiting to be performed, stations waiting for connection, or messages waiting for transmission.

read-after-write verification – Feature that protects data from failures in network hardware. When the network operating system writes data to a block on the hard disk, it reads back the data and compares it to the original data still in memory. If the data from the disk matches the data in memory, the data in memory is released. If the data does not match, Hot Fix marks that block on the disk as bad and redirects the data to another location on the hard disk.

real-time – An on-line computer that generates output nearly simultaneously with the corresponding inputs. Often, a computer system whose outputs follow its inputs by only a very short delay.

record locking – This feature on the network operating system prevents two users from writing simultaneously to the same record.

redundancy – A duplicate capacity that can be called upon when a failure occurs; having more than one path to a signal point.

remote dialback – Dials number back to confirm user’s number. It is a security method procedure.

rights – Security feature. Rights control which directories and files a user can access and what the user is allowed to do with those directories and files. Rights are assigned to directories and files by the network supervisor.

router – A software and hardware connection between two or more networks, usually of similar design, that permits traffic to be routed from one network to another on the basis of the intended destinations of that traffic. A router, formerly known as an internal or external bridge, can connect networks that use different network adapters or transmission media as long as both sides of the connection use the same protocols. If a router is located in a server, it is called an internal router; if located in a workstation, it is called an external router.

Sequenced Packet Exchange (SPX) – A protocol by which two workstations or applications communicate across the network. SPX uses IPX to deliver the messages, but SPX guarantees delivery of the messages and maintains the order of messages on the packet stream.

server – A computer on the network capable of recognizing and responding to client requests for services. These services can range from basic file and print services to support for complex, distributed applications. For example, a distributed database management system can create a single logical database across multiple servers.

Service Advertising Protocol (SAP) – A feature that advertises the services available on the Applications Server.

s hielded cable – Cable that has a layer of insulation to reduce electromagnetic interface.

spanning – A technique for improving I/O performance by placing frequently-used segments of a file system or database on separate disks.

standalone – A computer that is not connected to a network.

store-and-forward – Message-switching technique in which messages are temporarily stored at intermediate points before being transmitted to the next destination.

striping – A technique for improving I/O performance by interleaving file systems or data bases across multiple disks.

supervisor – The person responsible for the administration and maintenance of a network or database. A supervisor has access rights to all volumes, directories and files.

synchronous – A data transmission mode in which synchronization is established for an entire block of data (message). See also asynchronous.

Synchronous Data Link Control (SDLC) – IBM-defined link-control protocol that is code-independent.

System Application Architecture (SAA) – A set of IBM-defined standardsdesigned to provide a consistent environment for programmers and users across a broad range of IBM equipment, including microcomputers, minicomputers and mainframes.

System Fault Tolerance (SFT) – Duplicating data on multiple storage devices so that if one storage device fails, the data is available from another device. There are several levels of hardware and software system fault tolerance. Each level of redundancy (duplication) decreases the possibility of data loss.

S ystem Network Architecture (SNA) – IBM network architecture, defined in terms of its functions, formats and protocols.

TCP/IP (Transmission Control Protocol/Internet Protocol) – A protocol suite and related applications developed for the U.S. Department of Defense in the 1970s and 1980s specifically to permit different types of computers to communicate and exchange information with one another. TCP/IP is currently mandated as an official U.S. Department of Defense protocol and is also widely used in the UNIX community.

telephony – Generic term to describe voice telecommunications.

Telnet – Protocol in the TCP/IP suite that governs character-oriented terminal traffic.

terminal – A device, usually equipped with a keyboard and display, capable of sending and receiving data over a communications link.

terminal emulation – Software that allows a microcomputer to function as a dumb terminal.

throughput – Net data transfer rate between an information source and an information destination.

topology – The physical layout of network components (cable, stations, gateways, hubs and so on). There are three basic interconnection topologies-star, ring and bus networks.

transaction – A specific delimited amount of processing, intended to be an indivisible action.

Transaction Tracking System (TTS) – Feature that protects data from failures in network hardware. TTS protects the integrity of databases by backing out of incomplete transactions that result from a failure in a network component.

transparent – Function that operates without being evident to the user.

twisted-pair wiring – Two wires, usually loosely spun around each other to help cancel out any induced noise in balanced circuits.

uninterruptible power supply (UPS) – A backup power unit that provides continuous power even when the normal power supply is interrupted.

UNIX – Operating system developed by AT&T Bell Laboratories. Allows a computer to handle multiple users and programs simultaneously.

UnixWare operating system (OS) – Novell’s UNIX operating system for 80xxx processors.

UPS monitoring – Feature that protects data from failures in network hardware. A third-party uninterruptible power supply (UPS) provides power to the server during power fluctuations and outages. The network operating system’s UPS monitoring feature monitors the status of the UPS attached to the server.

user – Any person who attaches to a server or host.

user accounts – Security feature. Each user on a network has a user account. This account determines what name the user uses to log in to the network, the groups the user belongs to and what trustee assignments the user has. User accounts are maintained by the network supervisor.

value-added process (VAP) – An application that runs on top of network operating systems. VAPs tie in with the network operating system so that print servers, archive servers and database servers can provide services without interfering with the network’s normal operation.

VAX – A Digital Equipment Corporation minicomputer.

vector graphics (images) – Graphics displayed using a technology that specifies how an image is represented by using vector notation; for example, a starting point, a length and the direction that the line to be drawn from the starting point is to take. From a mathematical perspective, a vector graphics image is simply a collection of individual lines or vectors.

vertical application – An application that is specific to one area of use. For example: an accounting application or a legal application.

virtual – Conceptual or appearing to be, rather than actually being.

VMS (Virtual Memory System) – Operating system for DEC VAXs.

wide area network (WAN) – A WAN is two or more LANs in separate geographic locations connected by a remote link.

Windowing Korn Shell (WKSH) – A UNIX tool for developing windowing applications that incorporates the MoOLIT toolkit. It provides a comprehensive prototyping facility for exercising the application early in the development cycle.

workgroup – Two or more individuals on a LAN who share files, databases and other resources.

workstation – Any individual personal computer that is connected to a network.

X11 – X Windows System, version 11.

X.25 – A CCITT standard that defines the communications protocol for access to packet-switched networks.

X.400 – Open Systems Interconnection (OSI) standard that defines how messages are to be encoded for the transmission of electronic mail and graphics between dissimilar computers and terminals; defines what is in an electronic address and what the electronic envelope should look like. The X.400 standards are a subset of, and conform to, the X.25 standard approved by the Consultative Committee for International Telegraphy and Telephony (CCITT).

X Windows – Standard set of display-handling routines developed at MIT for UNIX workstations; they allow the creation of hardware-independent graphical user interfaces.

zone – On a local area network such as AppleTalk, a subgroup of users within a larger group of interconnected networks.