DIGITAL TCP/IP Services for OpenVMS
Concepts and Planning

congestion: The condition in which a network or part of a network is overloaded and has insufficient communication resources for the volume of traffic.

connection: A logical communication path between two processes that are using the TCP protocol. The communication path must exist before data can be sent in either direction. A three-way handshake occurs between the requesting and receiving process to establish a port through which the two processes communicate.

connection-oriented: The model of interconnection that consists of three phases: establish connection, transfer data, and release connection. TCP is a connection-oriented protocol.

connectionless: The model of interconnection in which communication takes place without first establishing a connection. UDP, IP, and IPX are connectionless protocols.

connectivity: The degree to which network nodes are interconnected. Full connectivity means all nodes have links to every other node.

container file: A data file on a DIGITAL UNIX NFS server with a UNIX directory structure and UNIX file attributes for a local, logical UNIX-style file system. Each UNIX regular file is stored as a separate data file using. The directory data files in the container file contain the UNIX file names and a pointer to the corresponding OpenVMS Files-11 data file.

Container File System (CFS): A logical UNIX-style file system that resides on a Files-11 formatted disk and is represented as a set of Files-11 files. See also container file.

contention: The condition when two or more stations attempt to use the same channel at the same time.

contention control: The scheme of access control used by many networks. Control is distributed among the nodes of the network. Any node wanting to transmit can do so, accessing the network on a first-come, first-served basis. However, it is possible that two nodes are in contention, or start transmitting at the same time, in which case a collision occurs. Each node must then back off and retransmit after waiting a random period of time.

control cluster: A group of small (256-byte) buffers dynamically allocated from nonpaged pool memory; stores information related to device sockets, internal control structures, IP addresses, Internet routes, and Internet packet headers.

Coordinated Universal Time (UTC): Greenwich Mean Time

cost: An OSPF (Open Shortest Path First) protocol metric. See metric and OSPF.

counters: The performance and error statistics kept for an entity by network management, such as lines and nodes.

CRC: See Cyclic Redundancy Check.

Cyclic Redundancy Check (CRC): An error detection scheme whereby a number is derived from a set of data before it is transmitted. Once transmitted, the receiving node recalculates the number and compares it to the value originally transmitted. If the numbers are different, some type of transmission error has occurred.

daemon: A process that executes in the background waiting for some event to occur.

data cluster: A group of large (1792-byte) buffers that store data in the system space; transmit and receive operations service user processes by moving data to and from data clusters.

Data Encryption Key (DEK): Used for encryption of message text and (with certain choices among a set of alternative algorithms) for computation of message integrity check (MIC) quantities.

Data Encryption Standard (DES): A type of encryption scheme approved by the U.S. National Bureau of Standards.

data link: A logical connection between two systems on the same circuit on which data integrity is maintained.

Data Link layer: The layer in a network model that handles communication between physical hosts.

data octet: See octet.

data overrun: The data blocks received that arrived too quickly to be processed by the receiver and were, therefore, lost.

datagram: A self-contained package of data carrying enough information to be routed from source to destination without reliance on earlier exchanges between source and destination or the transporting network.

datagram fragment: The result of fragmenting a datagram. Fragments carry a portion of data from the larger original and a copy of the original datagram header. The header fragmentation fields are adjusted to indicate the fragment's relative position within the original datagram.

datagram reassembly time: The time allowed for reassembly of a fragmented datagram.

datagram service: The mode of delivery for a datagram which is delivered in such a way that the receiver can determine the boundaries of the datagram as it was entered by the source.

DCE: See Distributed Computing Environment.

DCL: See Digital Command Language.

decision: The routing process that determines the path, or route, along which a data packet travels to reach its destination; forwards packets on the lowest-cost path even if that one does not have the fewest hops. The path that the data takes through the network is transparent to users.

decoding: The process by which the transfer syntax representation of a data value is transformed into the local representation of that value.

dedicated serial connection: A permanent connection between two hosts using an RS232 serial port. SLIP or PPP can be used for TCP/IP communication between the two hosts.

default route: The route used to direct any data addressed to network host addresses for which no explicit route is specified.

delay: A HELLO metric. Valid values are from 0 to 30000, inclusive. The value of 30000 is the maximum metric and means unreachable. See metric and HELLO.

delete access: The access right that grants users the ability to remove data from the domain.

DEK: See data encryption key.

DES: See Data Encryption Standard.

designated router: In OSPF, a designated router is a multiaccess network that has at least two attached routers. The designated router generates a link state advertisement for the multiaccess network and assists in running the protocol. The designated router is elected by the HELLO protocol.

destination address: The IP address that specifies where a datagram is to be sent; contains the network and host identifiers.

Any network or host.

destination port: A 2-octet value in the TCP and UDP header field that identifies the destination upper-level protocol for a packet's data.

device driver: The software associated with each physical device; serves as the interface between the operating system and the device controller.

device socket: The extension of the pseudodevice, used for communications; consists of the Internet pseudodevice and the socket. See also pseudodevice.

DHCP: See Dynamic Host Configuration Protocol

dialogue: The sequence of message exchanges between open systems that represents a single association and the set of underlying connections.

dialup: A temporary (as opposed to dedicated) network connection established through a telephone line with a modem.

dialup provider: A host that responds to incoming PPP connection requests. A PPP server.

Digital Command Language (DCL): The command interface of the OpenVMS operating system.

DIGITAL TCP/IP Services for OpenVMS: The Digital Equipment Corporation software product implemented on OpenVMS as an ancillary control process (ACP) and a network device driver (BG driver) with executive-level components and user applications that use TCP/IP protocols.

distance: An EGP metric. See metric and EGP. Valid values are from 0 to 255 inclusive.

Distributed Computing Environment (DCE): An architecture of standard programming interfaces, conventions, and server functions (for example, naming, distributed file system, remote procedure call) for transparently distributing applications across networks of heterogeneous computers.

distributed database: A collection of several different data repositories that look like a single database to the user. The Domain Name Service (DNS) is a distributed database.

distributed management: A form of network management in which network managers and management software are dispersed across many systems.

distributed processing: The technology that enables the distribution throughout the network of computing power and storage facilities to user work areas, such as offices, laboratories, or machines on factory floors.

distributed system: A collection of computer systems, tied together by communications networks for the purpose of sharing resources; end users do not need to be aware of the physical location of the shared resources.

DNS: See Domain Name Service.

domain: An organizational unit with administrative responsibility for naming networks or hosts. An internet domain name consists of a sequence of names (labels) separated by periods (dots); for example, tundra.mpk.ca.us.

domain name: The name used to refer to a fully qualified domain or subdomain. For example, in cat.food.iams.com, food.iams.com, iams.com, and .com are all domain names. Each name specifies a different domain level.

Domain Name Service (DNS): A distributed database system that allows TCP/IP applications to resolve a host name into a correct IP address.

dot address: See dotted decimal notation.

dotted decimal notation: The syntactic representation for a 32-bit integer that consists of four 8-bit numbers written in base 10 with periods (dots) separating them; used to represent IP addresses in the Internet, as in: 192.67.67.20. Many Internet application programs accept dotted decimal notation in place of destination machine names.

downline loading: Transferring a copy of a system image from a load host to a target. Some systems, such as DEC WANrouter systems and DECserver terminal servers, automatically request a downline load of their image upon startup and reboot. One of the functions of a TFTP server.

drift: The change in a clock's time rate over a specified period.

A measure, in hertz per second, of how quickly the skew of a clock is changing. See also skew.

dynamic adaptive routing: The automatic rerouting of traffic based on a sensing and analysis of current actual network conditions; not including cases of routing decisions taken on predefined information.

Dynamic Host Configuration Protocol: The Dynamic Host Configuration Protocol (DHCP), a superset of the BOOTP protocol, enables the automatic assignment of IP addresses to clients on networks from a pool of addresses. The IP address assignment and configuration occurs automatically whenever appropriate client systems (workstations and portable computers) attach to a network. The TCP/IP Services for OpenVMS implementation of DHCP is based on the JOIN product by Competitive Automation.

dynamic routing: A type of routing where a host or router talks to adjacent routers to learn what networks each router is connected to. Subsequently, the kernel's routing tables are updated when the router learns new information. There are many routing protocols including Interior Gateway Protocols (RIP, OSPF) and Exterior Gateway Protocols (EGP and BGP).

ephemeral port number: A port number temporarily assigned to a client process for the duration of a session. When the client process terminates, the port number can be assigned to another process. The port number is usually from 1024 to 5000.

EGP: See Exterior Gateway Protocol.

elective protocol: The classification in Internet standards for optional protocols.

electronic mail: The service whereby a computer user can exchange messages with other computer users (or groups of users) by means of a communications network; one of the most popular uses of the Internet.

e-mail: See electronic mail.

encapsulation: A technique used by layered protocols in which a layer adds header information to the protocol data unit (PDU) from the layer below. As an example, in Internet terminology, a packet would contain a header from the physical layer, followed by a header from the Network layer (IP), followed by a header from the Transport layer (TCP), followed by the application protocol data.

encryption: A process of encoding information so the meaning of its content is no longer immediately obvious to anyone who obtains a copy of it.

end node: See end system.

end system: A nonrouting system; can receive data packets addressed to it and send data packets to other systems on the same subnet but cannot relay, route, or forward data packets to other systems.

entity: An individual, manageable piece of a network; has attributes that describe it, a name that identifies it, and an interface that supports management operations.

entity class: A collection of entities that share the same properties and have the same parent entity; each member of the class has a unique identifier within the class. Entity classes have class names.

entity group: An architecturally defined collection of entities. The entities in the group must have a common top entity and must all be of the same class.

entity hierarchy: A logical hierarchical tree structures of manageable entities in which child entities are below their parent entities. Children can be accessed only through their parents' agent.

entity identifier: An attribute that specifically identifies an entity. See also attribute group.

entity name: A label associated with some entities used to identify or locate them for management purposes.

entity type: The subgrouping of an entity that determines its relationship to other entities.

Ethernet: A baseband network medium. Commonly used to connect a local area network.

event: A measurable network- or system-specific occurrence for which a logging component maintains a record.

experimental protocol: The classification in Internet standards for protocols that are developed as part of an ongoing research project not related to an operational service offering; not intended for operational use.

export database: The DIGITAL TCP/IP Services for OpenVMS database with directory names that can be mounted from remote NFS clients.

exported file: A file in an exported directory or a subdirectory of an exported directory. See also exporting.

exported file systems: A file system that can be accessed by a remote system using the Network File System. The local system imports the remote file system. Both the remote and local system must be configured to grant and receive access to the file system.

exporting: Identifying a directory on an NFS server that can be remotely mounted by NFS clients.

Extended File Specifications: A feature of the OpenVMS Alpha Version 7.2 that allows the use of Windows-style file specifications. This feature provides greater flexibility for OpenVMS Alpha systems to store, manage and access files that have names similar to those in a Windows 95 or Windows NT environment.

extended LAN: Multiple LANs connected with data link relays or bridges.

Exterior Gateway Protocol (EGP): The protocol that distributes routing information to the gateways that interconnect networks.

A class of routing protocols used to exchange routing information within an autonomous system.

One of a class of exterior gateway protocols.

FDDI: See Fiber Distributed Data Interface.

fetch/store operation: The operation of two commands that allow a system manager to fetch a value from a data item or to store a value into a data item.

Fiber Distributed Data Interface (FDDI): The high-speed (100 mb/s) networking standard based on fiber optics, established by the American National Standards Institute (ANSI); uses 1300 nanometer light wavelength. FDDI networks are limited to approximately 200 km in length, with repeaters every 2 km or less.

file: A uniquely named collection of information with shared managerial and structural properties.

file attribute: The characteristic of a file, such as its size or creation date. The values of some file attributes may change during the lifetime of a file.

file data: The information that is stored within a file and comprises its contents (as opposed to its attributes).

file designation: System-specific information that identifies a file on its storage system.

file server: The host whose principal purpose is to store files and provide network access to them.

file specification: System-specific information that identifies a file on its storage system.

file system: A method for recording, cataloging, and accessing files on a volume.

File Transfer Protocol (FTP): The protocol and software that permit a user on one host to access and transfer files to and from another host over a network. See also Trivial File Transport Protocol.

Files-11 ODS level 2 structure: The set of rules that govern the organization of the OpenVMS file system, external to the files themselves.

Finger utility: The utility that provides information about users on local and remote systems.

flow control: (1) The function of a receiving entity to limit the amount or rate of data that is sent by a transmitting entity. (2) The control of the rate at which hosts or gateways inject packets into a network or Internet, usually to avoid congestion. Flow control mechanisms can be implemented at various levels and allow communicating layers to match their data transfer and receive rates. Simplistic schemes, like ICMP source quench, simply ask the sender to cease transmission until congestion ends. More complex schemes vary the transmission rate continuously.

forwarder server: The name server that processes recursive requests that a slave server cannot resolve locally; has access to the Internet. See also BIND server, cache server, primary server, secondary server, and slave server.

forwarding information base: The table that GATED uses internally to store routing information it learns from routing protocols is a routing table; also known as a routing information base, or RIB. The routing table is used to collect and store routes from various protocols.

forwarding table: The table in the kernel that controls the forwarding of packets is a forwarding table, also known as a forwarding information base, or FIB.

FQDN: See fully qualified domain name.

fragment: A piece of a packet that results from a router dividing an IP datagram into smaller pieces for transmission across a network that cannot handle the original datagram size. Fragments use the same format as datagrams; fields in the IP header declare whether a datagram is a fragment and, if so, where the data in the fragment occurred in the original datagram. IP software at the receiving end must reassemble the fragments. See also maximum transmission unit.

fragmentation: The IP process of breaking up packets into smaller packets for transmission; allows a packet originating in a network that allows a large packet size to traverse a network that limits packets to a smaller size. The destination host reassembles the fragments. See also maximum transmission unit.

frame: A Data Link layer packet that contains the header and trailer information required by the physical medium

FTP: See File Transfer Protocol.

full-duplex circuit: A circuit designed for transmission in both directions at the same time. Contrast with half-duplex circuit.

full-duplex transmission: Data transmission in both directions at the same time. Contrast with half-duplex transmission.

fully qualified domain name (FQDN): The full site name of a system such as warren.enet.dec.com, rather than just its host name ---warren.

function code: A parameter in a $QIO system service call that defines the specific function of that $QIO.

GATED: A routing daemon that can be configured to route one or more of the following protocols: RIP, BGP, EGP, and OSPF.

gateway: A communications device or program that passes data between networks having similar functions but dissimilar implementations. The term "router" is now used in place of the original definition of "gateway."

An intermediate destination by which packets are delivered to their ultimate destination.

A host address of another router that is directly reachable through an attached network. As with any host address it may be specified symbolically.

gateway client: Another term for an access system.

gateway routing daemon: See GATED

GID: See group identification.

gigabit: One billion bits

gigabyte: One billion bytes

group identification (GID): The identification code for a group of UNIX users.

half-duplex circuit: A circuit designed for transmission in either direction, but only one direction at one time. Contrast with full-duplex circuit.

half-duplex transmission: Data transmission in either direction, but only one direction at a time. Contrast with full-duplex transmission.

handshaking sequence: The exchange of connection information between two communicating entities; takes place to enable the successful completion of a connection. Used, for example, in establishing a TCP connection between client and server applications.

hardware address: The address that identifies the connection device between the network controller of a host and the network cable. See also address.

hard link: A mechanism that allows you to assign more than one name to a file. Both the new name and the file being linked must be in the same file system. See link.

header: The portion of a packet that precedes the actual data and contains control information such as source and destination address and error checking.

header compression: A technique used by PPP and SLIP protocols to reduce the number of bytes per frame when sending packets over a slow serial link. The use of header compression is negotiated between the client and servers processes to reduce the size of the IP and TCP headers.

heterogeneous network: A network consisting of different network protocols or different operating system software, such as OpenVMS and UNIX.

hierarchical routing: Routing based on domains. Interdomain routers are responsible only for getting data to the right domain and intradomain routers take responsibility for routing within the domain.

hop count: The number of connections between two hosts, based on the number of different routers needed to traverse the distance between the two hosts.

hop: A term used in routing. Number of hosts separating a source and final destination (including the final destination) on a network.

host: A computer system that acts as a source or destination of network messages sometimes called "node."

The IP address of any system is usually specified in dotted-decimal notation. There are four values in the range from 0 to 255, inclusive, separated by dots (.). For example, 132.236.199.63 or 10.0.0.51. It can also be specified as an 8-digit hexadecimal string preceded by 0x. For example, 0x0a000043. In addition, if the options noresolv statement is not specified, this can be a symbolic host name. For example, gated.cornell.edu or nic.ddn.mil. The numeric forms are preferred over the symbolic form.

host address: See host number.

hosts database: The DIGITAL TCP/IP Services for OpenVMS database that is created by default; allows users to use host names; contains host names, IP addresses of the hosts, and any alias names for the hosts.

host name: The name given to a network host. See also fully qualified domain name and alias.

host number: The part of an IP address that identifies which host on the network is being addressed.

Host-to-Host Communication layer: Also called Transport layer. The second-highest level in the Internet architecture model; provides end-to-end communication services, including mechanisms such as end-to-end reliability and network control. Transmission Control Protocol (TCP) and User Datagram Protocol (UDP) reside in this layer.

IAB: See Internet Architecture Board.

IBM TN3270: The TELNET options that allow TELNET users to connect to hosts that support 3270 terminals.

ICMP: See Internet Control Message Protocol.

IETF: The Internet Engineering Task Force. A large international community of network designers, operators, vendors and researchers concerned with the evolution of the Internet architecture and the smooth operation of the Internet. Membership is open to everyone. See http://www.ietf.org/ for more information.

IGP: See Interior Gateway Protocol.

imported file: A file within a local NFS server that has been copied or linked to a remote NFS client.

inetd: A UNIX internet daemon. A server process listens for client requests for specific services. When inetd receives a request for a service, it starts the appropriate server process.