Nokia NRS II key terms

The mechanism by which an interior gateway protocol such as IS-IS or OSPF computes a loop-free tree using link metrics, so every router forwards toward a destination along the lowest accumulated cost path.

The two-tier hierarchy of Intermediate System to Intermediate System routing. Level 1 routers share a single area's topology; Level 2 routers form the backbone that connects areas and carries inter-area prefixes.

The Network Entity Title that uniquely identifies an IS-IS router. It is an OSI network service access point address, typically formatted as area identifier plus system identifier plus selector byte.

A logical grouping of OSPF routers that share a synchronized link-state database. Area 0 is the backbone, and all other areas must attach to it through an area border router to exchange summaries.

A vendor-assigned preference value that ranks routing sources when several protocols offer a route to the same prefix. Lower values win, and the losing route is kept as a standby in the routing table.

The controlled injection of routes learned by one routing protocol into another, requiring metric translation, filtering, and loop prevention because the receiving protocol cannot verify the original topology.

The ordered tie-breaking process a router applies to candidate BGP routes for one prefix, considering weight, local preference, AS path length, origin, MED, and finally eBGP over iBGP preference.

A well-known discretionary BGP attribute propagated only within an autonomous system. The highest value is preferred and is the standard tool for choosing an outbound exit point for the whole AS.

The Multi-Exit Discriminator, a non-transitive BGP attribute used to hint to a neighboring autonomous system which of several interconnection points is preferred for inbound traffic. Lower values are preferred.

A BGP attribute listing the autonomous systems a route has traversed. It provides loop detection and is the primary length-based tie-breaker between routes of equal local preference.

An internal BGP router allowed to reflect routes learned from one client or peer to other clients, removing the need for a full mesh of iBGP sessions inside a large autonomous system.

The forwarding rule that, among all routes containing a destination address, the entry with the most specific mask is used, regardless of protocol preference or metric.

The Routing Information Base holds all candidate routes from every protocol, while the Forwarding Information Base holds only the selected best routes installed in hardware for packet lookup.

A source-routing architecture in which the ingress node encodes a list of segment identifiers in the packet header, so core routers forward on the segment list instead of per-hop routing state.

The ordered set of MPLS labels that represents segment routing instructions in a packet. The top label is processed first, and each label may map to a node or an adjacency segment.

Topology-Independent Loop-Free Alternate, a local repair technique that precomputes a backup path using segment routing or MPLS labels so traffic is redirected within tens of milliseconds after a link failure.

Forwarding based on a short fixed-length label rather than the IP header, using a label forwarding table built by protocols such as LDP or RSVP-TE to create traffic-engineered paths.

An emulated point-to-point circuit that carries a native service such as Ethernet or TDM across a packet-switched network by encapsulating frames in MPLS or IP tunnels.

Virtual Private LAN Service, a multipoint Layer 2 VPN in which the provider network behaves as a single Ethernet switch, learning MAC addresses and flooding unknown frames to all participating sites.

Virtual Private Routed Network, a Layer 3 VPN in which each customer site connects to a provider edge router holding a per-customer VRF, and routes are exchanged as VPN-IPv4 or VPN-IPv6 addresses.

Virtual Routing and Forwarding, a separate routing table and forwarding instance on a router that lets overlapping customer address spaces coexist on shared physical infrastructure.

A BGP extended community attached to VPN routes that controls which VRFs import them. It decouples the topology of who receives a route from the address family itself.

An eight-byte prefix prepended to a customer prefix so that identical addresses from different VRFs become unique VPN-IPv4 routes inside the provider's multiprotocol BGP.

A loop-prevention rule that stops a router from advertising a route back out the interface or pseudowire from which it was learned, commonly applied in VPLS and distance-vector designs.

A Layer 2 protocol that elects a root bridge and blocks redundant links to produce a loop-free forwarding topology, converging after topology changes through BPDU exchange.

Multiple Spanning Tree Protocol, defined in IEEE 802.1s, which maps groups of VLANs to independent spanning tree instances so that different topologies can be active simultaneously.

Bonding several physical links into one logical bundle using LACP or static configuration, increasing bandwidth and providing sub-second resilience when a member link fails.

Multi-Chassis Link Aggregation, where two independent switches present a single LAG to an attached device, allowing active-active forwarding without relying solely on spanning tree blocking.

Encapsulating customer VLAN tags inside a provider VLAN tag (QinQ) so that many customers can share one service instance while keeping their own tag spaces independent.

The addressing scheme that replaced fixed address classes with variable-length prefixes, enabling route aggregation and a hierarchical allocation of IPv4 address space across providers.

A 128-bit addressing scheme with hexadecimal notation, a global routing prefix, subnet identifier, and interface identifier, plus link-local and multicast scopes used for neighbor discovery.

The ICMPv6-based protocol that resolves link-layer addresses, detects unreachable neighbors, and discovers routers and prefixes on an IPv6 link, replacing ARP and router discovery.

A function that forwards client broadcast requests to a DHCP server on another subnet, inserting the receiving interface address as the gateway address so the server can allocate correctly.

A DHCPv6 mechanism in which a provider router assigns a whole IPv6 prefix to a customer router, which then subnets it internally for its own LANs.

Assigning the same address to multiple nodes so that routing delivers each packet to the topologically nearest instance, commonly used for DNS resolvers and service endpoints.

A set of match and action statements, expressed as policy statements or route maps, that filters or modifies routes during import, export, or redistribution between protocols.

A management protocol and its data modeling language. NETCONF carries configuration and state as structured XML or JSON, while YANG defines the schema and constraints of that data.

The gRPC Network Management Interface, a streaming protocol for retrieving, setting, and subscribing to device state using YANG-modeled paths, widely used for telemetry and automation.

Continuous streaming of operational state from network devices, pushed on a schedule or on change to a collector, replacing periodic polling with near real-time, structured data.

An agentless automation tool that applies declarative playbooks over SSH or network APIs to configure many devices consistently, using inventory, modules, and idempotent tasks.

Managing network and compute configuration through version-controlled files and automated pipelines, so changes are reviewable, repeatable, and recoverable rather than applied manually.

Bringing a new device into service without manual configuration, using DHCP options, a bootstrap server, and signed configuration to load software and initial settings automatically.

A filter and rate limiter applied to traffic destined for the router's own CPU, protecting routing protocols and management access from floods and DoS attacks.

Unicast Reverse Path Forwarding, an antispoofing check that drops packets whose source address does not match the interface or route the router would use to reach it.

Restricting device administration through AAA, encrypted protocols such as SSH and SNMPv3, access lists, and role-based accounts, while disabling unused services and default credentials.

IEEE 802.1AE link-layer encryption and integrity protection between adjacent Ethernet peers, securing point-to-point links against eavesdropping and tampering without affecting upper-layer protocols.