Cisco SPCOR 350-501 key terms

A source-routing architecture in which the ingress node places an ordered list of segments in the packet header, and each segment identifies a node or adjacency the packet must traverse, removing per-hop state from core routers.

Segment routing implemented by encoding the segment list as a stack of MPLS labels. The label stack is pushed at the ingress and popped as segments are completed along the path.

Segment routing that encodes segments as 128-bit IPv6 addresses inside an IPv6 routing header, so no MPLS label stack is needed and segments can carry arguments or behaviors.

The value that names a single segment in a segment list, such as an MPLS label or an IPv6 address. Each SID maps to a forwarding instruction on the node that owns it.

A segment identifier assigned to a prefix so that any router advertising that prefix can forward traffic to it using the SID. It normally uses the node's loopback and represents shortest-path reachability.

A segment identifier bound to a specific neighbor link rather than a prefix, so a segment list can force traffic across one particular hop regardless of normal shortest-path selection.

Topology-Independent Loop-Free Alternate. A fast-reroute method that precomputes a backup path using a segment list, giving sub-50-millisecond protection for any topology without relying on physical topology constraints.

BGP Labeled Unicast, an address family in which BGP distributes prefixes together with an MPLS label. It is commonly used to extend labeled reachability between areas or autonomous systems.

Multiprotocol BGP, an extension that lets one BGP session carry multiple address families such as IPv4 unicast, IPv6 unicast, VPNv4, and labeled unicast, each negotiated separately.

A BGP router that receives routes from clients and reflects them to other clients, removing the need for a full iBGP mesh while preserving loop prevention through cluster and originator identifiers.

Label Distribution Protocol, used by routers to exchange label bindings for IGP-learned prefixes so that an MPLS label-switched path follows the IGP shortest path.

Resource Reservation Protocol with Traffic Engineering extensions. It signals explicit label-switched paths with bandwidth reservations and can reroute around failures using FRR.

Label-Switched Path, the ordered sequence of label-switching routers and labels that a packet follows through an MPLS network from ingress to egress.

Label Information Base, the control-plane table that stores label bindings learned from LDP, RSVP-TE, or BGP before the best binding is programmed into the forwarding table.

Label Forwarding Information Base, the data-plane table that maps an incoming label and interface to an outgoing label and next hop, driving actual MPLS packet forwarding.

Penultimate Hop Popping, in which the router one hop before the egress removes the top label so the egress receives a plain IP packet and avoids a second lookup.

A label value of 3 advertised by a router to request that its upstream neighbor pop the label before forwarding, the mechanism that implements PHP.

A traffic-engineered label-switched path built with RSVP-TE that follows an explicit or constrained path and can reserve bandwidth, separate from the IGP shortest path.

A protection mechanism that pre-installs a backup path or local repair so traffic is redirected within tens of milliseconds after a link or node failure, before IGP reconvergence completes.

A service in which a provider carries customer Layer 3 routing over a shared MPLS core using VRFs and MP-BGP VPNv4 or VPNv6 routes, keeping customer address spaces separate.

Virtual Routing and Forwarding, a per-customer routing table and forwarding instance on a PE router that isolates overlapping address space between different VPN customers.

An 8-byte value prepended to a customer prefix so identical prefixes from different VRFs become unique VPNv4 routes in BGP. It identifies the route but does not control import.

A BGP extended community attached to VPN routes that controls which VRFs import or export them, defining VPN membership and topology rather than mere uniqueness.

A controller-based WAN architecture with centralized management and policy, using overlay tunnels such as IPsec across any transport and application-aware routing at the edges.

The Cisco SD-WAN control protocol that runs between controllers and WAN edges, distributing routes, policies, and tunnel keys so edges can build the data-plane overlay.

The Cisco SD-WAN controller that holds the network's policy and distributes routes and policies to WAN edges over OMP, acting as the control-plane brain of the overlay.

Transport Locator, the Cisco SD-WAN identifier for a WAN edge's attachment to a particular transport, formed from system IP, color, and encapsulation, used to steer overlay tunnels.

A Cisco SD-WAN feature that continuously measures tunnel characteristics such as latency, loss, and jitter, then steers each application's traffic to the path that meets its SLA.

Ethernet VPN, a BGP address family that distributes MAC and IP reachability for Layer 2 and Layer 3 services, enabling multihoming, flooding control, and integrated routing and bridging.

A tunneling protocol that encapsulates Layer 2 frames in UDP and uses a 24-bit VNI, allowing far more segments than VLAN IDs and stretching Layer 2 across an IP underlay.

A design in which every leaf switch uses the same virtual MAC and IP for a distributed gateway, so a host's default gateway is always local no matter where it moves.

Quality of Service, the set of classification, marking, queuing, and scheduling mechanisms that give different treatment to traffic classes so latency and loss targets are met.

Differentiated Services Code Point, the 6-bit field in the IP header used to mark a packet's per-hop behavior class, such as expedited forwarding or assured forwarding.

Two rate-limiting techniques: policing drops or remarks traffic that exceeds a rate, while shaping buffers excess traffic and delays it to smooth bursts toward the committed rate.

A network management protocol that runs over SSH and uses YANG-modeled XML data for configuration and state retrieval, supporting candidate configs and atomic commit.

A data modeling language that defines the structure, types, and constraints of configuration and operational data for NETCONF, RESTCONF, and other model-driven interfaces.

A lightweight management protocol that exposes YANG-modeled data over HTTPS using HTTP verbs such as GET, POST, PUT, and DELETE, with JSON or XML payloads.

A push-based monitoring approach in which devices stream YANG-modeled operational data at a configured cadence to a collector, replacing frequent polling with near-real-time visibility.

gRPC Network Management Interface, a protocol that uses gRPC and protobuf encoding to get, set, and subscribe to YANG-modeled device state for high-frequency streaming telemetry.

A Cisco flow-export feature in which administrators define the flow keys and fields to record, then export cached flow records to a collector for traffic analysis and accounting.