Dell Networking Associate (D-NWG-FN-23) key terms

A frame-forwarding method where the switch receives the entire frame, checks its CRC for errors, and only then forwards it. It adds latency but prevents corrupt frames from propagating across the network.

A forwarding mode where the switch begins transmitting a frame as soon as it reads the destination MAC address, without waiting for the full frame. It minimizes latency but can forward errored frames.

A 32-bit cyclic redundancy check field appended to an Ethernet frame that lets the receiver detect bit errors introduced in transit. Switches discard frames whose FCS fails.

A Layer 2 control protocol that builds a loop-free logical topology by electing a root bridge and blocking redundant paths. It prevents broadcast storms when switches are cabled in redundant loops.

The STP control message exchanged between switches. It carries bridge ID, root path cost, port ID, and timers, allowing switches to agree on the loop-free topology and detect topology changes.

The process by which Spanning Tree switches choose a single reference switch, based on the lowest bridge priority and then the lowest MAC address. All path costs are computed relative to this root.

An evolution of STP defined in IEEE 802.1w that converges in seconds rather than tens of seconds by using explicit port roles and a proposal-agreement handshake instead of passive timer expiry.

An IEEE 802.3ad protocol that negotiates and maintains bundled links between two devices, verifying that both ends agree on which member ports form the aggregate and detecting misconfigurations.

A logical interface formed by bundling multiple physical links between the same pair of devices. It increases aggregate bandwidth and provides link-level redundancy without relying on Spanning Tree.

A technology that lets two separate switches present one logical port channel to a downstream device, so the server or access switch gains active-active dual-homed connectivity without a Spanning Tree block.

A Dell data center feature that pairs two switches into a single logical node using an inter-switch link and peer-link, providing active-active Layer 2 multipathing and faster convergence for attached servers.

The dedicated interconnect between two VLT switches that carries control traffic, synchronizes forwarding state, and handles traffic when the VLTi link set fails. It must be sized and cabled carefully to avoid split-brain.

A Layer 2 broadcast domain created logically on a switch so that ports on the same physical device can be isolated from one another. Frames are tagged with a 12-bit VLAN identifier on trunk links.

An IEEE standard that inserts a four-byte tag into Ethernet frames to carry VLAN membership across a shared link. Trunk ports carry multiple VLANs between switches or between a switch and a hypervisor.

The single VLAN whose frames travel untagged on an 802.1Q trunk. Both ends of the trunk must agree on it, otherwise traffic is misclassified and VLAN-hopping risks appear.

A switch interface assigned to exactly one VLAN that sends and receives untagged frames, typically connecting to an end host, printer, or server NIC that is unaware of VLAN tagging.

A per-switch database mapping learned source MAC addresses to the ports and VLANs they were seen on. It drives unicast forwarding decisions and is populated by source-address learning.

The behavior of a switch forwarding frames whose destination MAC address is not yet in the MAC address table out of every port in the VLAN, which is normal but can be abused or amplified.

A runaway condition in which broadcast or multicast frames multiply through a Layer 2 loop, saturating links and CPU. Spanning Tree, storm control, and loop detection are the usual countermeasures.

The control-plane table holding all routes a router has learned from static configuration and dynamic protocols, along with their administrative distance and metrics. The best routes are installed into the forwarding table.

The data-plane table derived from the best routes in the RIB. It is consulted for every packet and is typically implemented in hardware so lookups occur at line rate.

A vendor-assigned preference value that ranks route sources on the same device, so a router prefers, for example, a directly connected route over a static route over an OSPF route. Lower values win.

A value computed by a routing protocol that expresses the cost of reaching a destination through a particular path. Metrics are only comparable among routes from the same protocol.

The forwarding rule that, among all routes matching a packet's destination address, selects the one with the most specific mask. A /24 beats a /16 for any address inside both.

The metric used by OSPF, by default derived from interface bandwidth as reference bandwidth divided by interface bandwidth. It accumulates along the path and drives shortest-path tree computation.

A logical grouping of OSPF interfaces that limits link-state flooding and allows summarization at area borders. All non-backbone areas must connect to area 0 through an area border router.

Advertising a single aggregate prefix that covers many more-specific networks, shrinking routing tables and confining topology changes. It requires contiguous address space and a matching mask.

An open standard that lets several routers share a virtual IP and MAC address so hosts keep a working default gateway if one router fails. The master router forwards; backups take over.

Any mechanism that removes the single default-gateway failure point for end hosts, including VRRP and vendor-specific protocols. It is typically paired with fast failure detection.

A lightweight protocol that rapidly detects failures on a path between two devices, independent of routing protocol timers, so OSPF or BGP can converge in sub-second time.

An ordered set of permit and deny rules matched against packet fields such as source, destination, protocol, and port. The first match wins and an implicit deny typically follows.

A switch feature that limits which and how many source MAC addresses may appear on an interface, optionally shutting the port or dropping offending traffic when a violation occurs.

An IEEE framework in which a supplicant authenticates to the network through a switch before being granted access, using EAP over LAN between client and authenticator and RADIUS to the server.

A client-server protocol that centralizes authentication, authorization, and accounting for network access. Switches act as clients, forwarding credentials to a central server and applying the returned policy.

A Cisco-originated AAA protocol that separates authentication, authorization, and accounting and encrypts the entire payload. It is commonly used to control administrative CLI access to network devices.

A method of restricting device administration by assigning users to roles with defined privilege levels, so operators can run monitoring commands but not change configuration.

A UDP-based protocol for polling and trapping device statistics and events. SNMPv3 adds authentication and encryption, and is the version to use where security matters.

An unsolicited notification a device sends to a management station when a defined event occurs, such as a link going down or a fan failing, rather than waiting to be polled.

A hierarchical database of managed objects on a device, each identified by an object identifier. Management tools read and write MIB objects to monitor and configure equipment.

A standard for shipping event messages from network devices to a central collector, each with a facility and severity level, so logs from many devices can be correlated and retained.

A packet-sampling technology that provides traffic visibility on high-speed interfaces by exporting sampled headers to a collector, giving statistics without the cost of full packet capture.

A switch feature that copies traffic from one or more source ports or VLANs to a destination port for analysis by a protocol analyzer or intrusion detection system.

An automation approach in which a new switch boots, discovers a controller or configuration server through DHCP options or LLDP, and downloads its full configuration without manual CLI entry.

An HTTP-based programmatic interface using standard verbs and JSON or XML payloads, allowing scripts and orchestration tools to read and change device state instead of using the CLI.

Managing network and compute configuration through version-controlled declarative files rather than manual commands, enabling repeatable deployments, peer review, and rollback.

The practice of tracking, versioning, and enforcing intended device configurations, often with automated drift detection so devices that deviate from the baseline are flagged and corrected.

A two-tier data center topology in which every leaf switch connects to every spine switch, giving predictable, uniform latency and easy east-west scaling through equal-cost multipathing.

A tunneling encapsulation that carries Layer 2 frames inside UDP/IP packets, extending VLAN-like segments across a routed Layer 3 fabric. A 24-bit VNI provides far more segments than 802.1Q.

The entity that encapsulates and decapsulates VXLAN traffic, mapping a local segment to a VNI and a remote VTEP IP address. VTEPs are usually hosted on leaf switches.

The installation of multiple next hops with identical cost for the same prefix, so traffic is load-shared across them. It is fundamental to spine-leaf bandwidth utilization.

The set of classification, marking, queuing, and scheduling mechanisms that give latency-sensitive or business-critical traffic preferential treatment when links are congested.

An IEEE 802.1AB protocol that advertises a device's identity, capabilities, and port information to directly connected neighbors, aiding topology discovery, troubleshooting, and automated provisioning.