Glossaries

A

• ARS

  ARS (Adaptive Routing and Switching, referred to as Adaptive Routing Switch in some AsterNOS configuration documentation) is a flowlet-based load balancing tech

• ASIC

  A chip custom-built for one task; in a switch the forwarding ASIC moves packets in hardware at line rate.

B

• Basic Routing

  Routing is the most fundamental element of a data communication network. Routing information is the path information that tells a device where to send a packet,

• BFD

  BFD (Bidirectional Forwarding Detection) is a unified, network-wide mechanism for quickly monitoring the forwarding connectivity of links or IP routes. Before B

• BGP

  BGP (Border Gateway Protocol) is a distance vector routing protocol that establishes route reachability between autonomous systems and selects the best path amo

• BGP EPE

  BGP EPE (BGP Egress Peer Engineering) is an SRv6 traffic-engineering capability that lets a controller explicitly select which EBGP peer, physical link, or grou

• BGP FARE

  BGP FARE (Full Adaptive Routing Ethernet) is a BGP-based routing extension that gives switches deterministic, bandwidth-aware load balancing across multiple pat

C

• CGNAT

  CGNAT (Carrier-Grade NAT), also known as Large-Scale NAT (LSN), is a type of network address translation used by Internet Service Providers (ISPs) to let large

D

• Data Center Switch

  A data center switch is a high-performance network device designed to interconnect servers, storage, and other networking equipment inside a data center at spee

• DCBX

  DCBX (Data Center Bridging Exchange) is a link-layer protocol, based on IEEE 802.1Qaz, purpose-built to keep Data Center Bridging (DCB) configuration consistent

• DCQCN

  DCQCN (Data Center Quantized Congestion Notification) is a hybrid congestion control algorithm designed for RoCEv2 networks, built to balance throughput and lat

• DHCP Failover

  DHCP Failover is a technology for improving the reliability and availability of DHCP (Dynamic Host Configuration Protocol) service. In a traditional DHCP deploy

• DNS

  DNS (Domain Name System) is one of the core services underpinning the Internet: a distributed database that maps human-readable domain names to the IP addresses

E

• ECMP

  ECMP (Equal-Cost Multi-Path) is a routing technique for improving both network performance and reliability by spreading traffic across multiple links or ports o

• EVPN Multihoming

  EVPN Multihoming solves the same basic problem as MC-LAG — connecting a host to two or more switches for redundancy and load balancing — but does it natively wi

G

• gNMI

  gNMI (gRPC Network Management Interface) is an open-source management protocol built on top of gRPC, published and maintained by OpenConfig, for querying state,

• gRPC

  gRPC (Google Remote Procedure Calls) is a high-performance, general-purpose open-source RPC framework released by Google in 2015. It's built on HTTP/2, defines

I

• IGMP Snooping

  IGMP Snooping (Internet Group Management Protocol Snooping) is a Layer 2 multicast protocol that switches use to manage and control multicast traffic on a local

• IP Multicast

  IP multicast is a method of transmitting data across a network that lets a sender deliver a single stream to many receivers at the same time, instead of being l

• IPFIX

  IPFIX (IP Flow Information Export) is a standard protocol for collecting and exporting network traffic information, built to provide a universal, scalable way t

• IPSG

  IPSG (IP Source Guard) is a defense mechanism against IP address spoofing. It checks whether a device sending traffic on a given VLAN interface is who it claims

• IPv6

  IPv6 (Internet Protocol Version 6) is the second-generation network-layer protocol standard, designed to replace IPv4. Its most obvious difference from IPv4 is

• IS-IS

  IS-IS (Intermediate System to Intermediate System) is a link-state dynamic routing protocol originally designed by the International Organization for Standardiz

L

• LLDP

  LLDP (Link Layer Discovery Protocol), standardized in IEEE 802.1ab, is a Layer 2 discovery protocol — a way for two directly connected devices to exchange infor

M

• MACsec

  MACsec (Media Access Control Security) is a link-layer encryption technology standardized by IEEE 802.1AE. It protects data confidentiality, integrity, and repl

• MC-LAG

  MC-LAG (Multi-Chassis Link Aggregation Group) is a mechanism for implementing link aggregation across two physically separate devices, instead of just across po

• MPLS

  MPLS (Multiprotocol Label Switching) is a backbone-network technology that brings connection-oriented label switching to what is otherwise a connectionless IP n

• MSTP

  MSTP (Multiple Spanning Tree Protocol), standardized in IEEE 802.1s, is a spanning tree protocol built on top of STP and RSTP. It keeps everything that made STP

N

• NETCONF

  NETCONF (Network Configuration Protocol) is an IETF standards-based protocol, first defined in RFC 3535 and later formalized in RFC 6241, for installing, manipu

O

• Open Networking

  Open networking is an approach to building network infrastructure by separating (disaggregating) hardware from software, and using open standards and open-sourc

• OSPF

  OSPF (Open Shortest Path First) is a link-state interior gateway protocol developed by the IETF for distributing routing information within a single autonomous

P

• PFC

  PFC (Priority-based Flow Control) is an Ethernet flow-control mechanism that prevents packet loss on specific traffic priorities when a switch port becomes cong

• PIM-ASM

  PIM (Protocol Independent Multicast) is a routing solution for IP multicast that, true to its name, doesn't care which unicast routing protocol built the routin

• PoE Switch

  A Power over Ethernet (PoE) switch over a single Ethernet cable — eliminating the need for separate power adapters or dedicated electrical outlets for connected

• Policy-Based Routing

  Policy-Based Routing is a routing mechanism that forwards packets according to custom-defined policies rather than purely by destination address. Ordinary routi

• PPPoE

  PPPoE (Point-to-Point Protocol over Ethernet), defined in RFC 2516, carries PPP (Point-to-Point Protocol) over an Ethernet network. Its purpose is to establish

• PTP

  PTP (Precision Time Protocol) is a protocol for achieving high-precision clock synchronization across a computer network, standardized as IEEE 1588. Its goal is

Q

• QinQ

  QinQ — short for "802.1Q in 802.1Q" — is a simple Layer 2 VPN technique built on top of the IEEE 802.1Q standard, the specification that defines VLAN tagging on

• QoS

  QoS (Quality of Service) is a set of techniques for controlling network latency, congestion, and packet loss so that the traffic that matters most gets treated

R

• RDMA

  RDMA (Remote Direct Memory Access) is a networking technology that allows one computer to directly read from or write to the memory of another computer — bypass

• RIP

  RIP (Routing Information Protocol) is a distance vector routing protocol that determines the best path to a destination by counting the number of router hops al

• RoCEv2

  RDMA over routable UDP/IP Ethernet (UDP 4791) for low-latency, low-CPU transfers in AI/HPC fabrics; depends on near-lossless Ethernet (PFC/ECN).

S

• sFlow

  sFlow (Sampled Flow) is a packet-sampling-based network traffic monitoring technology, used primarily for statistical analysis of network traffic so operators c

• SNMP

  SNMP (Simple Network Management Protocol) is a standard, application-layer protocol used across the Internet for managing network devices. Network administrator

• SONiC

  SONiC decouples the network operating system from switch hardware. Built on Linux with a containerized architecture, it programs switching ASICs through SAI (th

• SR-IOV

  PCIe feature that splits one physical NIC into many virtual functions, letting VMs and containers bypass the hypervisor for near-native I/O.

• STP

  STP (Spanning Tree Protocol) is a Layer 2 management protocol that eliminates loops in switched Ethernet networks by selectively blocking redundant links, while

T

• Telemetry

  Telemetry, in networking, is technology for high-speed remote data collection from network devices. Instead of a management system repeatedly asking each device

• Teralynx 10

  Marvell 51.2 Tbps programmable switch ASIC for 800G AI/ML fabrics—ultra-low latency, high radix, 200MB+ buffers.

• Teralynx 7

  Marvell 3.2–12.8 Tbps programmable switch ASIC for 100G–400G leaf/spine—~500 ns latency, large buffers, AI/HPC fabrics.

V

• VPP

  User-space framework that forwards packets in cache-friendly vectors (batches) on commodity CPUs; used to build software routers and dataplanes.

• VRF

  VRF (Virtual Routing and Forwarding), also known as a VPN instance, is a way of logically dividing a single physical device into multiple independent virtual ro

• VRRP

  VRRP (Virtual Router Redundancy Protocol) is a standard protocol used to create a virtual gateway.

• VXLAN

  VXLAN (Virtual Extensible LAN) is a network virtualization technology that lets a Layer 2 network be built as an overlay on top of an existing Layer 3 network,

Z

• ZTP

  A switch self-configures on first boot by fetching its IP, OS image and config from network servers — no manual CLI needed.