Glossary

A specific implementation of 1000 Mbit/s operating over two multimode fiber cables, as specified in Standard IEEE 802.3z/D.50-1998.

A specific implementation of 1000 Mbit/s operating over four-pair Cat-5 or Cat-5e UTP cables, as specified in IEEE 802.3ab standards.

IEEE Standard supplement, now incorporated in IEEE 802.1D. Defines 8 priority levels for traffic classification at the data link level and suggests how they might be used.

IEEE Standard that specifies the architecture for VLAN tagging, association, and VLAN-capable bridges.

Name of the interface that corresponds to a given Internet address on a system.

A mechanism defined in IEEE 802.3u-1995 whereby devices sharing a link segment can exchange data and automatically configure themselves to operate at the highest capability mode shared between them. This is also used for link configuration per IEEE 802.3z and IEEE 802.3ab standards of duplex and flow control configuration between two 1000Base-SX/T links.

A number that uniquely identifies a device within a class. A class of devices is a logical grouping of similar devices.

A field in the message packet format identifying the end node(s) to which the packet is being sent.

Dynamic Host a Configuration Protocol

Data Link Provider Interface. An industry-standard definition for message communications to STREAMS-based network interface drivers.

A 10 Mbit/s LAN, developed by Digital Equipment Corporation, Intel and Xerox Corporation, upon which the IEEE 802.3 network is based.

A commonly used name applied to 100Base-T.

A mode of media utilization whereby data can flow in both directions simultaneously across the multiple wire pairs of a physical link. While full-duplex operation is not defined per se in the IEEE 802.3u-1995 specification, the specification does define a mechanism for this mode to be autonegotiated between devices on each end of a link. Full-duplex mode is typically found on switches.

The media utilization mode of IEEE 802.3u-1995 networks whereby data can flow in only one direction at a time across the multiple wire pairs of a physical link.

An identifier assigned by the system according to the physical location (slot) of the card in the hardware backplane. On HP servers, the I/O subsystem identifies each La card by its hardware path.

Name of system on the network.

A network interconnection device that allows multiple devices to share a single logical link segment.

A 10- or 100- Mbit/s LAN, specified in the IEEE 802.3u-1995 Standard for LANs. It uses the Carrier Sense Multiple Access/Collision Detection (CSMA/CD) network access method to give every node equal access to the network.

The Institute of Electrical and Electronics Engineers. A national association whose activities include publishing standards applicable to various electronic technologies. The IEEE technical committees are numbered and grouped by area. The 802.3 committee produced the standard for LAN networks.

The network address of a computer node. This address identifies both which network the host is on and which host it is. See the Installing and Administering LAN/9000 Software manual for detailed information about network addressing.

See Internet Address.

A software module that you can use to find and collect information about the Ethernet links on a system. It is based on the CIM (Common Information Model) standard from the distributed management task force (DMTF).

See Local Area Network.

A data communications system that allows a number of independent devices to communicate with each other.

The network to which a node is directly attached.

Unique value that identifies an individual hardware device.

Largest amount of data that can be transmitted through that interface. This value does not include the LLC or MAC headers.

Megabits per second

Megabytes per second

Management information base. An SNMP data structure that specifies what data can be obtained from or controlled in a device.

See Maximum Transmission Unit

A communication path through which messages can be sent and received. A hardware network interface has a hardware device associated with it, such as a LAN or FDDI card. A software network interface does not include a hardware device, for example, the loopback interface. For every IP address instance, there must be one network interface configured.

Any point in a network where services are provided or communications channels are interconnected. A node could be a workstation or a server processor.

The ability of a PCI-X I/O card to be added or replaced without the need to shut down or reboot the system. For instructions on how to perform OL*, see Configuring HP-UX For Peripherals.

A sequence of binary digits that is transmitted as a unit in a computer network. A packet usually contains control information plus data.

Peripheral Component Interconnect-Extended) An enhanced version of PCI bus technology originally developed by IBM, HP and Compaq. PCI-X is a superset of PCI; PCI-X is backward compatible with existing PCI cards. PCI-X cards can run in PCI slots though at the slower PCI speed. 64-bit PCI-X slots are longer than 32-bit PCI-X slots.

A unique number assigned to each network interface.

Path MTU

See Physical Point of Attachment

A specification for coding messages exchanged between two communications processes.

The name for the connector type used with UTP cabling.

Simple Network Management Protocol. A widely used network monitoring and control protocol. It uses devices or software processes called SNMP agents to return information about network devices. The information is collected into a MIB (Management Information Base).

A 32-bit mask which, when AND'd with an internet address, determines a subnetwork address. When the internet address is AND'd with the subnet mask, the ones in the host portion of the subnet mask will “overwrite” the corresponding bits of the host portion of the internet address, resulting in the subnet address. See Installing and Administering LAN Software for detailed information about subnet masks.

Small discrete physical networks connected via gateways which share the same network address space. See Installing and Administering LAN Software for detailed information about subnetworks and subnet addressing.Simple Network Management Protocol) A widely used network monitoring and control protocol. Data are passed from SNMP agents, which are hardware and/or software processes reporting activity in each network device (hub, router, bridge, etc.) to the workstation console used to oversee the network. The agents return information contained in a MIB (Management Information Base),

A network interconnection device that allows multiple connected senders and receivers to communicate simultaneously in contrast to a hub (repeater) where only one device can send at a time. Some switches have fixed port speeds (10 or 100 Mbit/s) while others allow port speeds to be configured or autonegotiated.

is a mechanism by which the host stack offloads certain portions of outbound TCP packet processing to the Network Interface Card (NIC) thereby reducing host CPU utilization.

The physical and logical geometry governing placement of nodes in a computer network. Also, the layout of the transmission medium for a network.

A data cable type consisting of pairs of wires twisted together without an electrically shielding jacket.

See Unshielded Twisted Pair

Virtual LAN. VLANs are a mechanism to determine which end stations should receive broadcast traffic, since it should not be sent arbitrarily to every connected user. Each packet transmitted by an end-station is assigned to a VLAN. An end-station only receives all the multicast and broadcast traffic on the LANs to which it belongs, and an end-station receives unicast traffic addressed to it on the VLAN to which it belongs.