With the plethora of devices, technologies and terminologies out there, keeping track of what all the different acronyms stand for and what they mean can be a headache. We all know what WWW and DHCP stand for, but this page is dedicated to some of the more esoteric and quirky acronyms that might be difficult to remember.
This list will be updated over time.
- TTFB – Time to first byte: Used commonly when referencing response times from things like browsers or remote hosts.
- ECMP –Equal Cost Multipath: Traffic load balancing across equal cost paths to the destination.
- IPAM – IP Address Management: System used to track IP assignments and allocations (typically a wiki or similar within an organisation)
- DCIM – Data Centre Infrastructure Management: Generic term that defines a system used to track and maintain Data Centre information, including everything from rack layout to change procedures.
- ISL – Inter-switch link: Link between switches.
- PHB – Per Hop Behaviour: How traffic is treated at each hop along its path. This term is typically used to define Differentiated Services QoS.
- PE – Provider Edge: Device on the edge of the Service Provider network (usually a router)
- CE – Customer Edge: Device on the edge of the Customer Network.
- CPE – Customer Premise Equipment: A networking device at the customers location. Sometimes used interchangeably with CE – however a CE device may not be a CPE (for example if the CE device is in a datacentre colo-location suite with the provider) and a CPE may not be a CE device (for example an internal switch that is not connected to the link between the provider and the customer).
- VLSM – Variable Length Subnet Mask: Generally use to refer to subnet masks that do not need to adhere to classful subnets.
- LFA – Loop Free Alternate: As per RFC 6571 a per-prefix LFA for a destination D at a node S is a pre-computed backup IGP next hop for that destination.
- DFZ – Default Free Zone: An area within a network where there is no default route, usually because the routers in question carry a full internet routing table.
- OOM – Out of Memory
Hardware and upgrades
- SMU – Software Maintenance Update
- PIE – Package Installation Envelope
- PSIRT – Product Security Incident Response Team
- DDTS – Distributed Defect Tracking System: The Cisco bug tracking system.
- FPD – Field Programmable Device: The firmware/software on the line cards that is separate from the OS running on the RP (Router Processor)
- NPE – Network Processing Engine: Routing engine for Cisco broadband router (frequently used in 7200 series routers)
- SPA – Shared Port Adapter
PPP (Point to Point Protocol)
- LCP – Link Control Protocol: First phase of PPP negotiation. LCP packets configure and test the data link. LCP packets have configuration options (authentication, maximum receive units etc).
- NCP – Network Control Protocol: Second phase, in which PPP choses and configures the network layer protocol (for example IPv4 or IPv6)
- IPCP – IP Control Protocol: NCP for IPv4.
- IPv6CP – IPv6 Control Protocol: NCP for IPv6.
- CMTS – Cable Modem Termination System: Cable modem CPE devices will connect to these in a typical cable architecture.
- DOCSIS – Data-over-Cable Service Interface Specification.
- MDT – Multicast Distribution Tree
- TIB – Tree Information Base
- PIM-SM -Protocol Independent Multicast, Sparse Mode
- PIM-DM -Protocol Independent Multicast, Dense Mode
- RPT – Rendezvous Point Trees: Used by PIM-SM and refers to a shared tree.
- SPT – Shortest Path Trees: Used by PIM-DM and refers to a source based tree.
- MSDP – Multicast Source Discovery Protocol: Used to discover multicast sources in other PIM domains.
- GLOP – Is not actually an acronym. Comes originally from RFC 2770 and is a globally assigned IPv4 multicast address that corresponds to a given ASN.
- OIL – Outgoing Interface List
- PIM-SSM – PIM Source Specific Multicast. Listeners specify what source they wish to get their multicast traffic from when joining a multicast group.
- MLD – Multicast Listener Discovery. Used by IPv6 in the same fashion that IGMP is used for IPv4.
- mRIB – Multicast RIB (Routing Information Base): Used to store multicast routes in the same way the RIB stores unicast routes.
- MOSPF – Multicast OSPF (Open Shortest Path First): An adapted version of OSPF for servicing multicast.
- GUA – Global Unicast Address
- ULA – Unique Local Address: Replaced Site Local Addressing for IPv6 since it has globally unique identifier in it.
- EUI-64 – Extended Unique Identifier-64: Used for auto-creation of the interface ID portion of an IPv6 address based on the MAC address.
- MIPv6 – Mobile IPv6: A protocol that allows nodes to remain reachable while moving around in the IPv6 Internet.
- SAS – Source Address Selection: IPv6 hosts can have multiple IPv6 Addresses on a given interface (Link-local, Global Unicast Address etc). SAS is an algorithm used to determine the source address to use when sending from such a device.
- DAS – Destination Address Selection: Algorithm used to determine the destination address to use when sending to an IPv6 device.
- NUD – Network Unreachability Detection
- DAD – Duplicate Address Detection
- DUID – DHCP Unique Identifier: How a DHCPv6 Server Identifies a host. There is exactly on of these for each host.
- IAID – Identity Associated Identifier: How to identify an interface in the DHCPv6 process.
- PD – Prefix Delegation: Mechanism to automate the assigning of IPv6 prefix blocks from ISP to customer (RFC 3633).
- RR – Requesting Router: In context of PD (Prefix Delegation), the RR is the router (usually the Customer Edge router) that makes the request to get an IPv6 prefix allocated.
- DR – Delegating Router: In the context of PD (Prefix Delegation), the DR is the router (usually the Provider Edge router) that delegates the IPv6 prefix to the RR (Requesting Router) after conferring with a AAA server.
- WKP – Well Known Prefix: Refers to the prefix 64:ff9b::/96. This prefix is used for multiple purposes, including stateful NAT64 on IOS-XE devices. This IPv4 address is usually embedded in the last 32 bits on the Well Known Prefix.
MPLS (Multiprotocol Label Switching)
- LSP – Label Switched Path: This is a series or LSRs (label switched routers) that switch a labelled packet through a part of an MPLS network. Not be confused with LSP with respect to IS-IS (see below).
- FEC – Forward Equivalence Class: As mentioned in RFC 3031, this refers to a group of packets, all of which are forwarded over the same path and with the same forwarding treatment.
- LSR – Label Switch Router: A router that is performing label switching.
- LER – Edge Label Switch Router or Label Edge Router: An ingress or egress LSR that typical moves packets from an IP forwarding into the MPLS label forwarding network. These are usually PE routers.
- VRF – Virtual Routing/Forwarding instance: A VRF within MPLS is associated with a customer VPN on a PE device. There is typical, among other tables and structures, one routing table for each VRF.
- TDP – Tag Distribution Protocol: Cisco’s proprietary control channel protocol used in MPLS frame-based networks to exchange label information.
- LDP – Label Distribution Protocol: An industry standard, non-proprietary control channel protocol used in MPLS frame-based networks to exchange label information.
- ARF – Automatic Route Filtering: The MPLS principle whereby an LER/PE router only accepts VPNv4 routes that are permitted by the locally configured VRFs.
- EOS – End of Stack: A bit in the header of an MPLS label indicating that it is the last label in the stack for that packet.
- LSD – Label Switch Database: Database structure where labels are stored (see LDP section below for more information).
- tLDP – Targeted LDP (Label Distribution Protocol): creates LDP neighborship between with devices that are not directly connected. Used for things like pseudowires (see LDP section below for more information).
- PW – Pseudowire: Within the context of MPLS, this a point-to-point connect across an MPLS network used in technologies like AToM (Any Transport over MPLS).
- LIB – Label Information Base: The data-structure or table within which label bindings are stored. This is usually populated by LDP, MP-BGP or RSVP (for TE).
LDP (Label Distribution Protocol)
- DoD – Downstream on Demand: an LDP label distribution method. In DoD an LSR will request a label for a given FEC from the downstream LSR as determined by the routing tables next-hop. Only LC-ATM uses DoD.
- UD – Unsolicited Downstream: another LDP label distribution method. Each LSE will send its label bindings to its LDP neighbors without being asked. This means that an LSR will typically receive multiple bindings for the same FEC (one from each neighboring LSR).
- LLR – Liberal Label Retention: an LDP label retention method. All received labels will be kept in the LIB. Only one will make it to the LFIB, but having unused labels in the LIB allows for faster convergence.
- CLR – Conservative Label Retention: another LDP label retention method: An LSR only stores on remote label binding – the one that corresponds to the next hop for a given FEC. LC-ATM uses this method. Less memory is used but convergence is slower.
IS-IS (Intermediate System to Intermediate System)
- LSP – Link state PDU (Protocol Data Unit): used by IS-IS to advertise routing information.
- IIH – IS-IS Hello: The Hello packet used in IS-IS (to detect neighbors, maintain adjacencies and so on).
- NSAP – Network Service Access Point: The type of addressing that IS-IS uses.
Peering and Transit
- PNI – Private Network Interconnect: Direct connection between only two networks for private peering.
- NPS – Network Policy Server.
- AAA – Authentication, Authorization and Accounting.
- RADIUS – Remote Authentication Dial-In User Service: Security protocol to provide AAA services. Encrypts the password field in the authentication only.
- TACACS – Terminal Access Controller Access-Control System: One of my favourites, simply for its repetition. This refers to a suite of Cisco proprietary security protocols for AAA services, the most well-known of which is TACACS+ (and is often what is referenced when TACACS is used). TACACS+ encrypts the entire payload in all off its AAA functions.
VxLAN (Virtual eXtensible Local Area Network)
- VNI – VxLAN Network Identifier: 24-bit segment ID. Used to identify and VxLAN network segment (layer 2 domain). Analogous to a VLAN ID.
- NVE – Network Virtualization Edge: This is an overlay interface that terminate a VxLAN tunnel.
- VTEP – VxLAN Tunnel Endpoint: As the name suggests, this is the terminating point for a VxLAN tunnel.
Bridging and Layer 2
- IRB – Integrated Routing and Bridging: Will need to be enabled on a router before bridge-domains can be configured.
- EFP – Ethernet Flow Point: This is an ethernet service endpoint. Allows Layer 2 flow decisions to be made within an interface.
- vPC – Virtual Port Channel: This is a form of multi-chassis aggregation. A cabled vPC peer link exists between the devices.
- mLACP – Multi-chassis Link Aggregation Control Protocol.
- ICCP – Inter-chassis Communication Protocol. Used in technologies like mLACP.
Registries and Global Routing
- RIR – Regional Internet Registry: These are registrars that have the ability to delegate prefixes and ASNs as well as create LIRs (Local Internet Registry, like an ISP). The follow are RIRs: AFRINIC, ARIN, APNIC, LACNIC and RIPE NCC.
- IRR – Internet Routing Registry: These can be thought of as databases that contain information on prefix and ASN assignments. They do not assign prefixes or ASNs themselves. RIRs can be IRRs as well. A list of IRRs can be found here.
- SFM – Switch Fabric Module: In the context of a 6500 series device, this provides the switching fabric between all fabric enabled modules. Used in conjunction with a Supervisor 1, 1a, 32 or 2. Must be installed in certain slots (in a 6509 for example, it must go in slot 5 or 6). A Sup720 supervisor module has the functionality of an SFM build into it.
- MSFC – Multilayer Switched Feature Card: Daughter card on a chassis-based switch that is installed on a supervisor module to enable, among other things, layer 3 functionality.
- PFC – Policy Feature Card: Daughter card on a chassis-based switch that is installed on a supervisor module to enable, among other things, QoS functions in hardware.
- PLIM – Physical Layer Interface Module: High performance CPU module that, amongst other tasks, performs carrier grade NAT.
PBB (Provider Backbone Bridging)
- BEB – Provider Backbone Edge Bridge: A PBB router on the edge of the provider network.
- BCB – Provider Backbone Core Bridge: A PBB router in the core of the provider network.
- SID – Service Identifier: Used to identify a PBB segment.
- S-VLAN – Service Provider VLAN: If an 802.1ad aggregation layer is used in the PBB environment, the S-VLAN refers to the outer 802.1Q tag allocated to frames in the aggregation layer, by the service provider.
- C-VLAN – Customer VLAN: Same principle as an S-VLAN (see above) but a C-VLAN will be assigned by and managed by the customer.
See this document for details on SSO, GR, NSF and NSR.
- SSO – Stateful Switchover
- GR – Graceful Restart
- NSF – Non-stop Forwarding
- NSR – Non-stop Routing
- DHD – Dual Home Device: A device with two uplinks.
- FHRP – First Hop Redundancy Protocol: A protocol to provide high availability to the gateway of a network.
- HSRP – Hot Standby Router Protocol: A Cisco proprietary FHRP.
- VRRP – Virtual Router Redundancy Protocol: An industry stand non-proprietary FHRP.
- BIA – Burned in address: Within the context of FHRPs, a virtual MAC address is typically used by the active, or primary, router to respond to ARP requests. To configure a router to respond with its burned in address means that it will respond to ARP requests with its connecting interfaces MAC address rather than the virtual MAC address.
BGP (Border Gateway Protocol)
- ORF – Outbound Route Filtering: Communicates with BGP peer to offload the route filter mechanism. The remote peer will apply a filter outbound rather than the local router apply an inbound filter.
- BPM – BGP Process Manager: This is a BGP process in IOS-XR. Among other things, it handles configuration changes made to BGP via the CLI. It checks for configuration errors before they are committed and also publishes the distribution of neighbors among the BGP speaker processes.
- bRIB – BGP RIB (Routing Information Base): This BGP process for IOS-XR calculates the final best path. There is a single instance of this process per address family.
- GTSM – Generalised TTL (Time to Live) Security Mechanism: This is not exclusive to BGP, but refers to RFC 5082 which proposes using a packet’s TTL field to protect the protocol stack against attack (typically CPU based attacks). In the context of BGP this refers to Cisco’s ttl security feature which ensures that eBGP neighbors are no more than a predefined number of hops away (usually just one hop) based on the TTL value.
- SOO – Site of Origin: A BGP attribute that is used to identify the router that redistributed a the prefix into BGP, in order to prevent routing loops.
Aggregation and Consolidation
- ICCP – Inter-Chassis Communication Protocol: Used to allow two separate devices to communicate and appear to other devices as one, for protocols like mLACP.
- mLACP – Multi-chassis Link Aggregation Protocol: Allows two devices to appear to a DHD as one device running LACP.
- MC-LAG – Multi-chassis Link Aggregation Group: A group of devices running mLACP.
- PoA – Point of Attachment: In the context of mLACP, a POA is an individual router within an MC-LAG.
- VSS – Virtual Switching System: The clustering of two 6500 chassis into one logical system.
SNMP (Simple Network Management Protocol)
- MIB – Management Information Base: This is a database of managed objects (represented by OIDs) that tracks agents through requests and traps.
- OID – Object Identifier: A reference that identified a managed object.
- NMS – Network Management Station: Usually a server that is used to monitor the network by means of polls to and traps from agents.
- SMI – Structure of Management Information: Standard that defines how managed objects are named and specifies their datatypes.
- ANS.1 – Abstract Syntax Notation 1: How data is represented and transmitted between agents and managers.
- BER – Basic Encoding Rules: Defines how managed objects are encoded for transmission.
- RMON – Remote Monitoring: Comes in versions 1 and 2. RMON helps provide the NMS (Network Management Station) with packet and/or network and application level statistics, through probes deployed on the network.
- USM – User-based Security Model: The security method used for SNMPv3
- VACM – View Access Control Model: Used by SNMPv3 to control access to managed objects in MIBs.
OSPF (Open Shortest Path First)
- LSA – Link State Advertisement: Data structure included in LSU (Link State Updates) that contain information about the network topology.They are stored in the LSDB (Link State Database).
- LSU – Link State Update: Packets sent between OSPF neighbors that contain topology information in the form of LSAs.
- ABR – Area Border Router: OSPF router connecting to more than one area.
- ASBR – Autonomous System Border Router: OSPF router running another routing protocol and redistributing into OSPF. It will send Type 5 LSAs into the OSPF domain.
- GRE – Generic Routing Encapsulation: A tunnelling protocol that encapsulates network layer technologies in point-to-point tunnels.
- mGRE – Multipoint GRE: GRE tunnelling that can connects to multiple end points.
- DMVPN – Dynamic Multipoint VPN (Virtual Private Network): An IP-over-IP point-to-multipoint tunnelling mechanism that using the dynamic establishing of mGRE tunnels.
- CAC – Call Admission Control: Used to decide whether or not a new voice or video call can be accepted based on several factors.
- LAC – L2TP Access Controller
- BRAS – Broadband Remote Access Server
- BNG – Broadband Network Gateway
- DSLAM – Digital Subscriber Line Access Multiplexer
QoS (Quality of Service)
- RED – Random Early Detection: A congestion avoidance mechanisms whereby packets are randomly dropped by a router prior to a queue becoming full. The intended effect is that the sending hosts who have had packets dropped will decrease their TCP window size and thus sending rate, thereby avoiding potential future congestion or tail-dropping.
- WRED – Weighted Random Early Detection: A variation of RED whereby the likelihood of a packet being dropped is dependent on the value of the IP Precedence or DSCP value in the packet header.
- ECN – Explicit Congestion Notification: A mechanism used to mark packet (usually by setting a bit in the header) in order to advise other devices on the network of congestion and adjust their sending rate accordingly.
- DSCP – Differentiated Services Code Point: A QoS field in an IP header used for marking and classifying traffic. Defined in RFC 2474.
- MQC – Modular QoS CLI: A common set of configuration commands used to setup QoS features on a Cisco router or switch.
- ToS – Type of Service: A byte in the an IP header whose bits are set using DSCP or IP Precedence standards for Quality of Service purposes.
- CS – Class Selector: A type of DSCP value used for compatibility with IP Precedence – typically just using the first three bits of the DSCP field of the ToS Byte.
- AS – Assured Forwarding: A type of DSCP value used to indicate both a class value and drop probability – e.g. AF21, where the class is 2 and drop probability is 1 (a higher drop probability indicating the packet is more likely to be dropped).
- ES – Express Forwarding: A DSCP value used to prioritise the packet by minimising delay, jitter and loss – but also using policing to ensure that EF marked packets do not prevent other classes of traffic from getting sufficient bandwidth.
- BE – Best Effort: This term generally refers to a class of traffic or behaviour whereby no QoS treatment is applied at all. In the case of markings, it indicates that the DSCP field, for example, will be all zeros.
- NBAR – Network Based Application Recognition: A classification and marking tool that enables inspection of packets at upper layers beyond 2-4 (MACs, IPs and port numbers) such as URL header content.
- PQ – Priority Queuing – A queuing mechanism whereby higher-priority queues are serviced before any lower-priority queues.
- CQ – Custom Queuing – A queuing mechanism designed to service all queues, even in times of congestion. Each queue is allocated a byte count. As packets are removed from a given queue by the scheduler, a counter increases until it reaches that queues byte count. Once the byte count is reached the scheduler will move to the next queue.
- FIFO – First In First Out – A queuing mechanism that forwards packets out in the order in which they were received. There is no scheduling logic used with just a single FIFO queue.
- WFQ – Weighted Fair Queuing – A queuing mechanism that classifies based on traffic flows rather than standard CB (Class Based) QoS that would use MQC tools.