CCNP Service Provider study materials
CCNP Service Provider study resources

CCNP Service Provider key terms

CCNP Service Provider · Get comfortable with the vocabulary.

Segment Routing

A source-routing architecture in which the ingress router encodes an ordered list of segments into the packet header, so core devices forward on that list instead of maintaining per-path state.

Related: SR-MPLS, Segment Identifier, SRGB
SR-MPLS

The instantiation of segment routing that carries segment lists as MPLS label stacks; each label represents a segment and is processed by existing MPLS forwarding hardware.

Related: Segment Routing, Segment Identifier, SRGB
Segment Identifier

A 32-bit value that names a segment in SR-MPLS, encoded as an MPLS label; it may represent a node, an adjacency, or a service path instruction.

Related: SR-MPLS, SRGB, Prefix-SID
SRGB

The Segment Routing Global Block is the range of local MPLS labels a router reserves for global segments; all nodes must use compatible ranges so prefix SIDs stay consistent.

Related: Segment Identifier, Prefix-SID, Segment Routing
Prefix-SID

A globally significant segment identifier assigned to a loopback prefix, so any node can steer traffic to that prefix by pushing the corresponding label.

Related: SRGB, Segment Identifier, Node SID
Anycast SID

A prefix SID advertised by multiple routers for the same address, letting traffic follow equal-cost paths to the nearest of those routers and providing inherent redundancy.

Related: Prefix-SID, Segment Routing
TI-LFA

Topology-Independent Loop-Free Alternate computes a backup path using segment lists that avoids micro-loops after a failure, giving sub-50-millisecond protection without relying on topology shape.

Related: Segment Routing, Fast Reroute, Prefix-SID
BGP-LU

BGP Labeled Unicast carries an MPLS label with each prefix advertisement, commonly used to distribute interior and inter-AS routes with labels in service provider cores.

Related: Route Reflector, MPLS
Route Reflector

A BGP speaker that receives routes from clients and reflects them to other clients, removing the need for a full iBGP mesh inside an autonomous system.

Related: BGP-LU, Route Target, Autonomous System
BGP Confederation

A method of splitting a large autonomous system into sub-ASes that run iBGP internally and eBGP-like sessions between sub-ASes, reducing full-mesh requirements while presenting one AS externally.

Related: Route Reflector, Autonomous System
Autonomous System

A set of routers under one administrative control that presents a single routing policy to the Internet, identified by a unique number assigned by a registry.

Related: BGP Confederation, Route Reflector
MPLS

Multiprotocol Label Switching forwards packets using short fixed-length labels instead of destination lookups; the label stack allows tunneling, traffic engineering, and VPN separation.

Related: LDP, LSP, Label Stack
LDP

The Label Distribution Protocol exchanges label bindings between adjacent routers for IGP-learned prefixes, building label-switched paths automatically without traffic-engineering constraints.

Related: MPLS, LSP, Label Stack
LSP

A Label Switched Path is the ordered sequence of routers and labels a packet traverses; it may be built by LDP, RSVP-TE, or segment routing.

Related: LDP, RSVP-TE, MPLS
Label Stack

The ordered set of MPLS labels on a packet; the top label drives forwarding at each hop while lower labels identify the VPN or service at the egress.

Related: MPLS, LSP, Penultimate Hop Popping
Penultimate Hop Popping

A behavior where the second-to-last router removes the top label before forwarding, so the egress router receives the packet with the service label already exposed.

Related: Label Stack, MPLS
RSVP-TE

An MPLS signaling protocol that reserves bandwidth and establishes explicitly routed label-switched paths, supporting traffic engineering and fast reroute through constraint-based path computation.

Related: LSP, MPLS, Fast Reroute
Fast Reroute

A protection mechanism that pre-computes and pre-installs a backup path so traffic is redirected locally within milliseconds of a link or node failure.

Related: RSVP-TE, TI-LFA
L3VPN

A service in which a provider carries customer IP routes over a shared MPLS backbone using VRFs and MP-BGP, isolating each customer's routing table from the others.

Related: VRF, Route Distinguisher, Route Target
VRF

A Virtual Routing and Forwarding instance is a separate routing and forwarding table on a provider edge router, one per customer or service, keeping overlapping addresses independent.

Related: L3VPN, Route Distinguisher
Route Distinguisher

An eight-byte value prepended to a customer prefix so that identical addresses from different VPNs become unique in MP-BGP; it does not control import or export.

Related: VRF, Route Target, L3VPN
Route Target

A BGP extended community attached to VPN routes that determines which VRFs import them, effectively defining VPN membership and topology.

Related: Route Distinguisher, VRF, L3VPN
MP-BGP

Multiprotocol BGP extends BGP with address-family identifiers so it can carry VPN routes, IPv6, and other families alongside traditional IPv4 unicast.

Related: L3VPN, Route Target
VPLS

Virtual Private LAN Service offers multipoint Layer 2 connectivity across an MPLS network, making remote sites appear on one bridged Ethernet segment.

Related: EVPN, Pseudowire
EVPN

Ethernet VPN uses BGP control-plane signaling to advertise MAC and IP reachability for Layer 2 and Layer 3 services, replacing flood-and-learn with programmable, multipoint-aware signaling.

Related: VPLS, Route Target, Pseudowire
Pseudowire

An emulated point-to-point circuit that carries a customer Layer 2 protocol across a packet-switched network, using control and encapsulation to preserve the original service.

Related: VPLS, EVPN, MPLS
VXLAN

A tunnel encapsulation that carries Layer 2 frames in UDP over an IP network, using a 24-bit virtual network identifier to scale tenant separation beyond VLAN limits.

Related: EVPN, Spine-Leaf
Spine-Leaf

A two-tier data center topology where every leaf connects to every spine, giving predictable latency and equal-cost multipathing; spines forward but do not host endpoints.

Related: VXLAN, ECMP
ECMP

Equal-Cost Multipath spreads traffic across several equally preferred next hops, increasing bandwidth utilization and resilience without changing routing metrics.

Related: Spine-Leaf, TI-LFA
IS-IS

A link-state interior gateway protocol that runs directly on Layer 2, scales well in large provider cores, and supports flexible TLVs used for traffic engineering and segment routing.

Related: OSPF, Segment Routing
OSPF

An interior gateway protocol that floods link-state advertisements to build a shortest-path tree; areas limit flooding scope and summarize routes at area boundaries.

Related: IS-IS, ECMP
Prefix Suppression

A technique that stops advertising transit link addresses so only loopbacks and edge prefixes remain in the IGP, shrinking the topology database and improving convergence.

Related: IS-IS, OSPF
QoS

Quality of Service classifies, marks, queues, and polices traffic so latency-sensitive and business-critical flows get appropriate treatment during congestion.

Related: H-QoS, Policing
H-QoS

Hierarchical QoS applies shaping and scheduling at two levels, such as per-subscriber and per-session, so providers can enforce both aggregate and individual service guarantees.

Related: QoS, Policing
Policing

A traffic conditioner that meters a flow against a committed rate and drops or remarks excess packets, keeping traffic within contract without buffering.

Related: QoS, Shaping
Shaping

A traffic conditioner that buffers excess packets and releases them at the contracted rate, smoothing bursts instead of dropping them as policing does.

Related: Policing, H-QoS
NETCONF

A network management protocol that uses XML-encoded remote procedure calls over a secure transport to read and change device configuration in a structured, transactional way.

Related: YANG, gNMI, Model-Driven Telemetry
YANG

A data modeling language that defines the structure, constraints, and semantics of device configuration and state, enabling vendor-neutral automation and validation.

Related: NETCONF, gNMI, Model-Driven Telemetry
gNMI

A gRPC-based protocol for retrieving, setting, and subscribing to device state using modeled paths, widely used for streaming telemetry and automation.

Related: YANG, Model-Driven Telemetry, NETCONF
Model-Driven Telemetry

A push model in which devices stream structured operational data on a schedule or on change, replacing periodic polling with near-real-time, high-resolution monitoring.

Related: gNMI, YANG, Streaming Telemetry
Streaming Telemetry

Continuous delivery of device metrics and counters to a collector over a subscription, giving finer granularity and lower overhead than SNMP polling.

Related: Model-Driven Telemetry, gNMI
SNMP

A legacy management protocol that polls MIB objects or receives traps, still common for basic fault monitoring but limited in scale and data granularity.

Related: Streaming Telemetry, Model-Driven Telemetry