Network Ing Authority

Networking Services Glossary: Key Terms and Definitions

Networking services span a broad and often overlapping set of technologies, contractual arrangements, and infrastructure patterns — each with precise technical meanings that differ significantly from casual usage. This glossary defines the core terms encountered across enterprise, small business, and carrier-grade networking contexts, drawing on definitions from the Internet Engineering Task Force (IETF), the National Institute of Standards and Technology (NIST), and the Institute of Electrical and Electronics Engineers (IEEE). Accurate terminology reduces procurement errors, aligns technical and business stakeholders, and supports compliance documentation for regulated industries.


Definition and Scope

A networking services glossary functions as a controlled vocabulary for the discipline of designing, procuring, operating, and auditing data communications infrastructure. The scope covers Layer 1 (physical) through Layer 7 (application) of the OSI Reference Model, encompassing both on-premises and cloud-delivered network functions. Terminology is drawn from three primary classification systems:

The IEEE 802 standards family governs local area network (LAN) specifications, while the IETF RFC corpus defines routing protocols, addressing schemes, and transport mechanisms that underpin wide-area and internet-facing services. NIST Special Publication 800-160 Vol. 1 provides systems engineering vocabulary applicable to network security architecture.

Terms in this glossary are not vendor-defined. Where a vendor uses a proprietary label (e.g., "SD-WAN Fabric," "Secure Access Cloud"), the underlying concept maps to an industry-standard definition provided here.


How It Works

Networking terminology operates within a layered definitional hierarchy. A single service — such as SD-WAN — can be described simultaneously at the transport layer (overlay tunneling), the management plane (centralized policy control), and the commercial layer (consumption-based billing). Understanding which definitional layer is in use prevents miscommunication between network engineers, procurement officers, and compliance teams.

The major definitional categories and their constituent terms are organized as follows:

  1. Physical infrastructure terms — fiber, copper, wireless spectrum, rack units (RU), patch panels, structured cabling (governed by ANSI/TIA-568 standards)
  2. Logical network terms — VLAN (IEEE 802.1Q), subnet, CIDR notation, routing table, BGP autonomous system number (ASN)
  3. Protocol and service terms — MPLS (Multiprotocol Label Switching), QoS (Quality of Service), SIP (Session Initiation Protocol, RFC 3261), DHCP (RFC 2131), DNS (RFC 1034/1035)
  4. Architecture and topology terms — hub-and-spoke, full mesh, spine-leaf, zero-trust architecture (NIST SP 800-207)
  5. Managed service and commercial terms — SLA (Service Level Agreement), SLO (Service Level Objective), NOC (Network Operations Center), MTTR (Mean Time to Repair), uptime percentage

Each category maps to a distinct layer of technical and contractual responsibility. A managed network services contract, for example, will reference SLA and MTTR terms from Category 5, while its technical specifications reference Categories 2 through 4.


Common Scenarios

Glossary misalignment produces measurable operational failures. The following scenarios illustrate where definitional precision is most consequential.

Scenario 1 — WAN procurement ambiguity. A business requesting "dedicated bandwidth" may receive either a true dedicated circuit (e.g., a DS3 at 44.736 Mbps, point-to-point) or a dedicated internet access (DIA) product with a committed information rate (CIR). These are technically distinct: dedicated circuits offer deterministic latency; DIA products do not guarantee hop-by-hop path quality. The WAN services reference elaborates on this distinction.

Scenario 2 — Security service classification. "Firewall as a service" (FWaaS), "network detection and response" (NDR), and "managed detection and response" (MDR) are frequently conflated in vendor proposals. FWaaS operates at the network perimeter (Layers 3–4 and optionally Layer 7); NDR applies behavioral analytics to internal traffic flows; MDR combines tooling with human analyst response. The network managed detection and response reference page defines the MDR boundary specifically.

Scenario 3 — Cloud networking terminology drift. Terms like "virtual private cloud" (VPC), "transit gateway," and "cloud-native load balancer" are cloud-provider-specific implementations of canonical networking concepts (private network segment, inter-AS routing, and Layer 4/7 load distribution respectively). Mapping provider terminology to OSI layers clarifies capability and interoperability requirements. Cloud networking services provides this mapping in detail.


Decision Boundaries

Selecting the correct term — and confirming that a vendor uses it consistently — requires applying four boundary tests:

  1. Layer boundary test — Identify which OSI layer(s) the service or component operates on. Two services described with the same label but operating on different layers are not interchangeable.
  2. Management plane boundary test — Distinguish data plane functions (packet forwarding) from control plane functions (routing decisions) and management plane functions (configuration, monitoring). IETF RFC 3654 defines these planes formally.
  3. Responsibility boundary test — Confirm whether a term describes a customer-managed, provider-managed, or shared-responsibility function. Network compliance and regulatory requirements addresses how regulatory frameworks assign these responsibilities.
  4. Contractual boundary test — Verify that SLA terms (uptime, MTTR, RTO, RPO) are quantified in the service agreement, not implied by marketing language. An "always-on" description carries no contractual weight unless it maps to a specific availability percentage (e.g., 99.99% equates to approximately 52.6 minutes of allowable downtime per year under standard calculation).

Managed vs. Unmanaged — The primary binary distinction in networking services vocabulary separates managed services (provider monitors, maintains, and remediates) from unmanaged or self-managed infrastructure (customer assumes operational responsibility post-provisioning). This distinction controls cost, staffing requirements, compliance documentation scope, and incident response timelines. Every term in this glossary can be prefixed with "managed" or "unmanaged" to produce a distinct service category with different SLA structures.


References

On this site

Core Topics
Contact

In the network