— Appendix B —
Glossary

Abstract Syntax Notation One (ASN.1)—An OSI language used to define datatypes for networks. It is used within TCP/IP to provide conformance with the OSI model.

Access Control—A process that defines each user's privileges on a system.

Acknowledgment (ACK)—A positive response returned from a receiver to the sender indicating success. TCP uses acknowledgments to indicate the successful reception of a packet.

Active Open—An operation performed by a client to establish a TCP connection with a server.

Address—A memory location in a particular machine's RAM. A numeric identifier or symbolic name that specifies the location of a particular machine or device on a network, and a means of identifying a complete network, subnetwork, or a node within a network.

Address Mask (also called the subnet mask)—A set of rules for omitting parts of a complete IP address in order to reach the target destination without using a broadcast message. The mask can, for example, indicate a subnetwork portion of a larger network. In TCP/IP, the address mask uses the 32-bit IP address.

Address Resolution—Mapping of an IP address to a machine's physical address. TCP/IP uses the Address Resolution Protocol (ARP) for this function.

Address Resolution Protocol (ARP)—See Address Resolution. ARP is a protocol used to correlate an IP address to a machine's physical address. The reverse operation is performed by Reverse Address Resolution protocol (RARP).

Address Space—A range of memory addresses available to an application program.

Advanced Research Project Agency (ARPA)—DARPA's former name. ARPA was an agency funded by the U.S. federal government originally for pure research. When it was changed to DARPA the funding became part of the Defence budget.

Agent—In TCP/IP, an agent is an SNMP process that responds to get and set requests. Agents can also send trap messages.

American National Standards Institute (ANSI)—The U.S. body responsible for setting standards.

Application Layer—The highest layer in the OSF model. It establishes communications rights and can initiate a connection between two applications.

Application Programming Interface (API)—A set of routines available to developers and applications to provide specific services used by the system, usually specific to the application's purpose. They act as access methods into the application.

ARPANET (Advanced Research Projects Agency Network)—A packet-switched network that later became known as the Internet.

ASCII (American National Standard Code for Information Interchange)—An 8-bit character set defining alphanumeric characters.

Assigned Numbers—Used in Request For Comment (RFC) documents to specify values used by TCP/IP.

Asynchronous—Communications without a regular time basis, enabling transmission at unequal rates.

Autonomous System—A collection of routers that are under the control of a single management body. The system usually uses a common Interior Gateway Protocol.

Backbone—A set of nodes and links connected together comprising a network, or the upper layer protocols used in a network. Sometimes the term is used to refer to a network's physical media.

Bandwidth—The range of frequencies transmitted on a channel, or the difference between the highest and lowest frequencies transmitted across a channel.

Baseband—A type of channel where data transmission is carried across only one communications channel, supporting only one signal transmission at a time. Ethernet is a baseband system.

Baseband Signaling—A type of transmission that has a continuous encoded signal. Only one node at a time can send data over this type of transmission technology. Used in Local Area Networks.

Basic Encoding Rules (BER)—The rules for encoding datatypes using ASN.1.

Baud—The number of times a signal changes state in one second.

Berkeley Software Distribution (BSD)—A version of the UNIX operating system that first included TCP/IP support. The UNIX operating systems that included TCP/IP are referred to as 4.2BSD or 4.3BSD.

Bit Error Rate (BER)—The number of errors expected in a transmission.

Bit Rate—The rate that bits are transmitted, usually expressed in seconds.

BITNET (Because It's Time Network)—An electronic mail network connecting over 200 universities. It merged with the CSNET network to produce CREN.

Block Mode—A string of data recorded or transmitted as a unit. Block mode transmission is usually used for high speed transmissions and in large, high speed networks.

Border Gateway Protocol (BGP)—A protocol that provides information about the devices that can be reached through a router (into an autonomous network). BGP is newer than EGP.

Bridge—A network device capable of connecting networks that use similar protocols.

Broadband (also known as wideband)—A range of frequencies divided into several narrower bands. Each band can be used for different purposes.

Broadband Signaling—The type of signaling used in Local Area Networks that enables multiplexing of more than one transmission at a time.

Broadcast—The simultaneous transmission of the same data to all nodes connected to the network.

Brouter—A network device that is a combination of the functions of a bridge and a router. It can function as a bridge while filtering protocols and packets destined for nodes on different networks.

BSD—See Berkeley Software Distribution.

Buffer—A memory area used for handling input and output.

Burst Mode—A transmission mode where data is transmitted in bursts rather than in continuous streams.

Bus—In network topology, a linear configuration. Also used to refer to part of the electronic layout of network devices.

Cache—A memory location that keeps frequently requested material ready. Usually the cache is faster than a storage device. It is used to speed data and instruction transfer.

Carrier Sense—A signal generated by the physical network layer to inform the data link layer that one or more nodes are transmitting on the network medium.

Carrier Sense Multiple Access with Collision Detection (CSMA/CD)—A network media access control protocol wherein a device listens to the medium to monitor traffic. If there is no signal, the device is allowed to send data.

Cheapernet—A reduced-cost Ethernet variant where the maximum length of the network is 200 feet. It uses inexpensive 75-ohm coaxial cable, simple connectors, and no transceivers.

Client—A program that tries to connect to another program (usually on another machine) called a server. The client calls the server. The server listens for calls.

Client/Server Architecture—A catch-all term used to refer to a distributed environment where one program can initiate a session and another program can answer its requests. The origin of client/server designs is closely allied with the TCP/IP protocol suite.

CMOT—The TCP/IP implementation of CMIP.

Collision—An event that occurs when two or more nodes broadcast packets at the same time—the packets collide.

Collision Detection—A device's capability to detect whether a collision has occurred.

Common Management Information Protocol (CMIP)—A network management protocol usually associated with OSI. When used with TCP/IP, CMIP is called CMOT.

Common Management Information Service (CMIS)—Management services provided by CMIP.

Connection—A link between two or more processes, applications, machines, network, and so forth. Connections can be logical, physical, or both.

Connection Oriented—A type of network service where the transport layer protocol sends acknowledgments to the sender regarding incoming data. This type of service usually provides for retransmission of corrupted or lost data.

Connectionless—A type of network service that does not send acknowledgments to the sender upon receipt of data. UDP is a connectionless protocol.

Consortium for Research and Education Network (CREN)—The name for the body arising from the combination of CSNET and BITNET.

Contention—A condition occurring in some LANs where the Media Access Control (MAC) sublayer allows more than one node to transmit at the same time, risking collisions.

CMIP—See Common Management Information Protocol.

CMOT—The TCP/IP implementation of CMIP.

Core Gateway—A router operated by the Internet Network Operations Center to distribute routing information.

Crosstalk—Signals that interfere with another signal.

CSNET (Computer Science Network)—An electronic mail network that merged with BITNET to form CREN.

Cyclic Redundancy Check (CRC)—A mathematical function performed on the contents of an entity that is then included to enable a receiving system to recalculate the value and compare to the original. If the values are different, corruption of the contents has occurred.

Daemon—A UNIX process that operates continuously and unattended to perform a service. TCP/IP uses several daemons to establish communications processes and provide server facilities.

DARPA (Defense Advanced Research Project Agency)—The governmental body that created the DARPANET for widespread communications. DARPANET eventually became the Internet.

Data Circuit-Terminating Equipment (D CE)—Required equipment to attach Data Terminal Equipment (DTE) to a network or serial line. A modem is a DCE device. Also called Data Communications Equipment and Data Circuit Equipment.

Data Encryption Standard (DES)—An encryption standard officially sanctioned in the U.S.

Data Link—The part of a node controlled by a data link protocol. It is the logical connection between two nodes.

Data Link Protocol (DLP)—A method of handling the establishment, maintenance, and termination of a logical link between nodes. Ethernet is one example of a Data Link Protocol.

Data Terminal Equipment (DTE)—The source or destination of data, usually attached to a network by DCE devices. A terminal or computer acting as a node on a network is usually a DTE device.

Datagram—A basic unit of data used with TCP/IP.

Defense Communications Agency (DCA)—The governmental agency responsible for the Defense Data Network (DDN).

Defense Data Network (DDN)—Refers to military networks such as MILNET and ARPANET and the communications protocols (including TCP/IP) that they employ.

Destination Address—The destination device's address.

Directory System Agent (DSA)—A program that accepts queries from a directory user agent (DUA).

Directory User Agent (DUA)—A program that helps a user to send a query to a directory server.

Distributed Computing Environment (DCE)—A set of technologies developed by the Open Software Foundation (0SF) supporting distributed computing.

Distributed File Service (DFS)—An Open Software Foundation (OSF) fileserver technology sometimes used with TCP/IP.

Distributed Management Environment (DME)—A system and network management technology developed by the Open Software Foundation (OSF).

Distributed Processing—When a process is spread over two or more devices, it is distributed. It is usually used to spread CPU loads among a network of machines.

Domain Name System (DNS)—A service that converts symbolic node names to IP addresses. DNS is frequently used with TCP/IP. DNS uses a distributed database.

Dotted Decimal Notation—A representation of IP addresses. Also called "dotted quad notation" because it uses four sets of numbers separated by decimals (for example, 255.255.255.255).

Double Byte Character Set—A character set where alphanumeric characters are represented by two bytes, instead of one byte as with ASCII. Double byte characters are often necessary for Asian languages, which have more than 255 symbols.

Drop Cable—In Ethernet networks it refers to the cable connecting the device to the network, sometimes through a transceiver.

Dumb Terminal—A terminal with no significant processing capability of its own, usually with no graphics capabilities beyond the ASCII set.

Emulation—A program that simulates another device. For example, a 3270 emulator emulates an IBM 3270 terminal, sending the same codes as the real device would.

Encapsulation—Including an incoming message into a larger message by adding information at the front, back, or both. Encapsulation is used by layered network protocols. With each layer, new headers and trailers are added.

Enterprise Network—A generic term usually referring to a Wide Area Network providing services to all of a corporation's sites.

Ethernet—A data link level protocol comprising the OSI model's bottom two layers. It is a broadcast networking technology that can use several different physical media, including twisted pair cable and coaxial cable. Ethernet usually uses CSMA/CD. TCP/IP is commonly used with Ethernet networks.

Ethernet Address—A 48-bit address commonly referred to as a physical or hard address that uniquely identifies the Ethernet Network Interface Card (NIC) and hence the device the card resides in.

Ethernet Meltdown—A slang term for a situation where an Ethernet network becomes saturated. The condition usually persists for only a short time and is usually caused by a misrouted or invalid packet.

Extended Binary Coded Decimal Interchange Code (EBCDIC)—An alternative to ASCII used extensively in IBM machinery. Some other vendors use it for mainframes. EBCDIC and ASCII are not compatible but are easy to convert between.

Exterior Gateway Protocol (EGP)—A protocol used by gateways to transfer information about devices that can be reached within their autonomous systems.

Fiber Distributed Data Interface (FDDI)—An ANSI-defined standard for high speed data transfer over fiber optic cabling.

File Server—A process that provides access to a file from remote devices. Also used to refer to the physical server itself, although the term server also implies other services than file provision in most client/server networks.

File Transfer Access Method (FTAM)—A file transfer program and protocol developed by OSI. It includes some basic management functions.

File Transfer Protocol (FTP)—A TCP/IP application used for transferring files from one system to another.

Fragmentation—The breaking of a datagram into several smaller pieces, usually because the original datagram was too large for the network or software.

Frame—Usually refers to the completed Ethernet packet, which includes the original data and all the TCP/IP layers' headers and trailers (including the Ethernet's).

Frame Check Sequence (FCS)—A mathematical function used to verify the integrity of bits in a frame, similar to the Cyclic Redundancy Check (CRC).

Frame Relay—A network switching mechanism for routing frames as quickly as possible.

Gateway—In Internet terms, a gateway is a device that routes datagrams. More recently used to refer to any networking device that translates protocols of one type network into those of another network.

Gateway-to-Gateway Protocol (GGP)—A protocol used to exchange routing information between core routers.

Gigabyte—One billion bytes, corresponding to decimal 1,073,741,824 (a kilobyte is 1,024 decimal).

Government Open System Interconnection Profile (GOSIP)—A government standard that uses the OSI reference model.

Hardware Address—The low-level address associated with each device on a network, usually corresponding to the unique identifier of the network interface card (NIC). Ethernet addresses are 48 bits.

High Level Data Link Control (HDLC)—An international data communication standard.

Hop Count—The number of bridges that data crosses in a Token Ring network.

IEEE 802.2—An IEEE-approved data link standard used with the 802.3, 802.4, and 802.5 protocol standards.

IEEE 802.3—An IEEE-approved physical layer standard that uses CSMA/CD on a bus network topology.

IEEE 802.4—An IEEE-approved physical layer standard that uses token passing on a bus network topology.

IEEE 802.5—An IEEE-approved physical layer standard that uses token passing on a ring network topology.

Initial Sequence Number (ISN)—A number defined during the startup of a connection using TCP. Used to number datagrams.

Initiate—In TCP/IP, to send a request for something (usually a connection).

Institute of Electrical and Electronic Engineers (IEEE)—A professional organization for engineers that also proposes and approves standards.

Integrated Service Digital Network (ISDN)—A set of standards for integrating multiple services (voice, data, video, and so on).

Interface—A shared point between two software applications or two hardware devices.

Interior Gateway Protocol (IGP)—A protocol used by gateways in an autonomous system to transfer routing information.

International Organization for Standardization (ISO)—An international body composed of individual countries' standards groups that focuses on international standards.

Internet—A collection of networks connected together that span the world that uses the NFSNET as its backbone. The Internet is the specific term for a more general internetwork or collection of networks.

Internet Activities Board (IAB)—The Internet group that coordinates the development of the TCP/IP protocol suite.

Internet Address—A 32-bit address used to identify hosts and networks on the Internet.

Internet Control Message Protocol (ICMP)—A control and error message protocol that works in conjunction with the Internet Protocol (IP).

Internet Engineering Notes (IEN)—Documents that discuss TCP/IP, available through the Network Information Center (NIC).

Internet Engineering Steering Group (IESG)—The executive party of the IETF.

Internet Engineering Task Force (IETF)—Part of the IAB responsible for short-term engineering needs relating to the TCP/IP protocol suite.

Internet Protocol (IP)—The part of TCP/IP that handles routing.

Internet Research Task Force (IRTF)—A part of the IAB that concentrates on research and development of the TCP/IP protocol suite.

IP Address—A 32-bit identifier that is unique to each network device.

IP Datagram—The basic unit of information passed through a TCP/IP network. The datagram header contains source and destination IP addresses.

IS-IS (Intermediate System to Intermediate System Protocol)—A routing protocol that performs routing functions with IP and OSI data.

ISO Reference Model—The seven-layer ISO networking model. It isolates specific functions within each layer.

ISODE (ISO Development Environment)—An attempt to develop software that enables OSI protocols to run on TCP/IP.

Jam—An Ethernet term for communicating with all devices on a network on which a collision has occurred.

Jitter—A term used with 10BaseT (twisted-pair Ethernet) networks where signals are out of phase with one another.

Kerberos—An authentication scheme developed at MIT used to prevent unauthorized monitoring of logins and passwords.

LAN (Local Area Network)—A collection of devices connected to enable communications between themselves on a single physical medium.

Learning Bridge—A network bridge device that has the function of a bridge and the capability to monitor the network in order to determine which nodes are connected to it, and adjust routing data accordingly.

Leased Line—A dedicated communication line between two points. Usually used by organizations to connect computers over a dedicated telephone circuit.

Link—A generic term referring to a connection between two end points.

Logical—Conveys an abstract concept in a simpler manner, such as using a logical machine name instead of its physical address.

Logical Link Control (LLC)—The upper part of the data link sublayer protocol that is responsible for governing the exchange of data between two end points.

Mail Exchanger—A system used to relay mail into a network.

Management Information Base (MIB)—A database used by SNMP containing configuration and statistical information about devices on a network.

Maximum Segment Size—The maximum permissible size for the data part of a packet.

Maximum Transmission Unit (MTU)—The largest datagram that can be handled by a specific network. The MTU can change over different networks, even if the transport is the same (such as Ethernet).

Media Access Control (MAC)—The lower half of the data link sublayer that is responsible for framing data and controlling the physical link between two end points.

Medium Access Unit (MAU)—A MAU handles the connection of a device operating on a network.

Message Transfer Agent (MTA)—A process that moves messages between devices.

Metropolitan Area Network (MAN)—An IEEE-approved network that supports high speeds over a metropolitan area.

MILNET (Military Network)—A network that was originally part of ARPANET, now designated for exclusive military use in installations that require reliable network services.

Modem (Modulator-Demodulator)—A device that converts digital signals into analog signals and vice versa. Used for conversion of signals for transmission over telephone lines.

Modem Eliminator—A device that functions as two modems to provide service for data terminal equipment (DTE) and data communications equipment (DCE).

Multihomed Host—A device attached to two or more networks.

Multiplex—Simultaneously transmitting multiple signals over one channel.

Name Resolution—The process of mapping aliases to an address. The Domain Name System (DNS) is one system that does this.

National Institute of Standards and Technology (NIST)—A U.S. standards body previously called the National Bureau of Standards that promotes communications-oriented standards.

National Research and Education Network (NREN)—A network backbone supporting large capacities planned for future Internet use.

National Science Foundation Network (NFSNET)—The network that acts as part of the Internet backbone.

NetBIOS (Network Basic Input/Output Operating System)—A network programming interface typically used to connect PCs together.

Network—A number of devices connected to enable the device to communicate with any other device over a physical medium.

Network Address—For TCP/IP, the 32-bit IP address of a device.

Network File System (NFS)—A protocol developed by Sun MicroSystems that enables clients to mount remote directories onto their own local filesystem.

Network Information Center (NIC)—The Internet administration facility that controls the naming of networks accessible over the Internet.

Network Information Service (NIS)—A set of protocols developed by Sun Microsystems used to provide directory services for network information.

Network Interface Card (NIC)—A generic term for a networking interface board used to connect a device to the network. The NIC is where the physical connection to the network occurs.

Network Management—Any aspect of monitoring or controlling a network, including all administration details.

Network Service Access Point—Used to identify an OSI device and point to the transport layer.

Network Virtual Terminal (NVT)—Protocols that govern virtual terminal emulation.

Node—A generic term used to refer to network devices.

Open Shortest Path First (OSPF)—The basic Internet routing protocol for sending data over multiple paths. It uses the network's topology for routing decisions.

Open Software Foundation (OSF)—A consortium of hardware and software vendors collaborating to produce technologies for device-independent operation.

Open Systems Interconnection (OSI)—A family of ISO-developed standards relating to data communications.

Optical Fiber—A plastic or glass cable that uses light as a communications medium.

Packet—In TCP/IP, a term referring to the data passing between the internet layer and the data link layer. Also a generic term used to refer to data transferred through a network.

Passive Open—An action taken by a server daemon to prepare it to receive requests from clients.

PING (Packet Internet Groper)—A utility program used to test a system's TCP/IP software by sending an ICMP echo request and then waiting for a response.

Point-to-Point—Transmission directly between two points without any intervening devices.

Point-to-Point Protocol (PPP)—A TCP/IP protocol that provides host-to-network and router-to-router connections. Can be used to provide a serial line connection between two machines.

Port—A number used to identify TCP/IP applications. Generally a port is an entry or exit point.

Protocol—Rules governing the behavior or method of operation of something.

Protocol Conversion—The process of changing one protocol to another.

Protocol Data Unit (PDU)—A term used in TCP/IP to refer to a unit of data, headers, and trailers at any layer in a network.

Protocol Interpreter (PI)—A process that carries out FTP functions. FTP uses one Protocol Interpreter for the server and another one for the user.

Proxy—A mechanism whereby one system functions for another when responding to protocol requests.

Push Service—A service provided by TCP to enable an application to specify when data must be transmitted as soon as possible.

RARP—See Reverse Address Resolution Protocol.

Receive Window—A range of sequence numbers that a sender can transmit at a given time.

Remote Network Monitor (RMON)—A device (such as a workstation) that collects and maintains information about network traffic.

Remote Procedure Call (RPC)—A TCP/IP protocol that provides a routine which calls a server that returns output and status (return) codes to the client.

Repeater—A network device that boosts the power of incoming signals to enable the length of a network to be extended.

Requests for Comment (RFCs)—Documents containing specifications for TCP/IP protocols. RFCs are also used to propose new protocols. RFCs are available from the Network Information Center (NIC).

Resolver—Software that enables clients to access the Domain Name System (DNS) database and acquire an address.

Resource—Usually refers to application programs, but also used generally to refer to system capabilities such as memory, networks, and so on.

Retransmission Timeout—Occurs when data has been sent to a destination but no acknowledgment has been received when a timer has expired. When a retransmission timeout occurs, the protocol usually resends the data.

Reverse Address Resolution Protocol (RARP)—A TCP/IP protocol that enables a device to acquire its IP address by performing a broadcast on the network.

RIP (Routing Information Protocol)—A TCP/IP protocol used to exchange information about routing. Usually used when only a small number of computers are in use.

rlogin—Remote login service that enables a user on one machine to log in as a user on another. It is similar to Telnet.

Round Trip Time—The time for a TCP segment to be sent and its acknowledgment received.

Router—A device that connects LANs into an internetwork and routes traffic between them.

Routing—The process of determining a path to use to send data to its destination.

Routing Information Protocol (RIP)—A protocol used to exchange information between routers.

Routing Table—A list of valid paths through which data can be transmitted.

RS232C—A physical layer specification for connecting devices. Commonly used for serial lines.

SAP (Service Access Point)—The location at which two applications can exchange information.

Segment—A protocol data unit (PDU) that consists of a TCP header and (optional) data. Also used to refer to parts of a network that is divided into smaller parts (segments).

Send Window—A range of sequence numbers that can be received.

Sequence Number—A 32-bit field in the IP header that identifies the datagram.

Serial—A sequence of events occurring one after another.

Serial Line Internet Protocol (SLIP)—A protocol used to utilize TCP/IP over serial lines.

Server—An application that answers requests from other devices (clients). Also used as a generic term for any device that provides services to the rest of the network, such as printing, high-capacity storage, and network access.

Simple Mail Transfer Protocol (SMTP)—In TCP/IP, an application providing electronic mail services.

Socket—In TCP/IP, an addressable point that consists of an IP address and a TCP or UDP port number that provides applications with access to TCP/IP protocols.

Socket Address—The complete designation of a TCP/IP node consisting of a 32-bit IP address and a 16-bit port number.

Socket Descriptor—An integer used by an application to identify the connection.

Source—The originating device.

Source Routing—A routing method determined by the source device.

Subnet—In TCP/IP, part of a TCP/IP network identified by a portion of the Internet address.

Subnet Address—The part of the IP address that identifies the subnetwork.

Subnet Mask—A set of bits that excludes networks from having a system-wide broadcast, instead restricting the broadcast to a subnetwork.

Switched Connection—A data link connection that is established on demand (like a telephone call).

SYN—A segment used in the start of a TCP connection to enable both devices to exchange information defining characteristics about the session. It is also used to synchronize the target and destination devices.

Synchronous Data Transfer—The transfer of data between two nodes at a timed rate (as opposed to asynchronously).

Telnet—A TCP/IP application that enables a user to log in to a remote device.

TCP/IP—Transmission Control Protocol/Internet Protocol.

10Base2—An Ethernet term meaning a maximum transfer rate of 10 Megabits per second that uses baseband signaling, with a contiguous cable segment length of 100 meters and a maximum of 2 segments.

10Base5—An Ethernet term meaning a maximum transfer rate of 10 Megabits per second that uses baseband signaling, with 5 continuous segments not exceeding 100 meters per segment.

lOBaseT—An Ethernet term meaning a maximum transfer rate of 10 Megabits per second that uses baseband signaling and twisted pair cabling.

Terminal Server—A network device that provides physical access for dumb terminals, usually using an abbreviated TCP/IP protocol to enable a dumb terminal to remotely log on.

Terminator—A resistor that must be on both ends of thick and thin Ethernet networks.

Throughput—The amount of data that can be transferred through a medium within a certain time period.

Time-to-Live (TTL)—The amount of time a datagram can remain on the internetwork. It is usually specified as the number of hops to permit.

Token Ring—A lower layer connection-based networking protocol using a token passing method to control data traffic.

Topology—The configuration of network devices.

Traffic—A general term used to describe the amount of data on a network backbone.

Transceiver—A network device required in baseband networks that takes a digital signal and puts it on the analog baseband medium. Transceivers can sense collisions.

Transmission Class 4—An OSI transport layer protocol similar to TCP. Often referred to as OSI TP4.

Transmission Control Block (TCB)—A data structure that holds information about TCP and UDP connections.

Transmission Control Protocol (TCP)—A transport layer protocol that is part of the TCP/IP protocol suite and provides a connection-based, reliable data stream.

Trivial File Transfer Protocol (TFTP)—A mechanism for remote logons similar to Telnet but that uses UDP as a transport layer protocol instead of TCP.

UDP (User Datagram Protocol)—A connectionless transport layer protocol. It does not perform retransmission of data.

User Agent—An electronic mail program that helps end users manage messages.

User Service—A service provided by TCP permitting an application to specify that data being transmitted is urgent and should be processed as soon as possible.

Well Known Port—In TCP/IP, an address for an agreed upon purpose.

Wide Area Network (WAN)—Usually used to refer to a network spanning large geographic distances.

X.400—A protocol defining standards for electronic mail in an open network.

X.500—A protocol defining standards for directory services in an open network.

X Series—A collection of widely accepted standards, including data communications.

X Windows—A software protocol developed at MIT for a distributed windowing system. X uses TCP for a transport protocol.

XNS (Xerox Networking Standard)—Networking protocols developed by Xerox, similar to TCP/IP.