Federated Network Information Model

Description

The Federated Network Information Model (FNIM) is a cornerstone of the 3GPP Management and Orchestration (MANO) framework, specifically within the Network Resource Model (NRM) family of specifications. It is not a single database but a standardized, object-oriented information model that defines how network elements, their capabilities, configurations, and relationships are represented as managed objects. This model provides a common language and structure for management data exchanged between different management systems, such as Element Managers (EMs), Network Management Systems (NMS), and higher-level orchestration systems like the Network Function Virtualization Orchestrator (NFVO).

The FNIM works by specifying a hierarchy of managed object classes, each with defined attributes, operations, and notifications. These classes model everything from physical hardware (e.g., equipment racks, cards, ports) to virtualized network functions (VNFs), logical connections, and performance measurements. A key architectural principle is federation: the FNIM allows for the integration of information from multiple, potentially heterogeneous management domains (e.g., a RAN domain and a core network domain, or networks from different operators) into a cohesive, unified view for an overarching management system. This is achieved through standardized reference points and information service interfaces that allow systems to discover, retrieve, and manipulate managed object instances based on the FNIM schema.

Key components of the FNIM include the Generic Network Resource Model (GRNM), which provides base object classes, and technology-specific derivatives that extend these classes for particular network domains like 5G NR, LTE, or 5GC network functions. The model supports full lifecycle management operations—creation, configuration, monitoring, and termination—of network resources. Its role is to act as the 'single source of truth' schema that enables automated provisioning, fault correlation across domains, service assurance, and closed-loop optimization. By adhering to the FNIM, management systems can interoperate without requiring custom, point-to-point integrations, which is essential for managing large-scale, multi-vendor 5G and cloud-native networks.

Purpose & Motivation

The FNIM was created to address the critical challenge of management complexity and fragmentation in modern telecommunications networks. Prior to its standardization, network management relied heavily on proprietary Management Information Bases (MIBs) and interfaces specific to each vendor's equipment. This made end-to-end service provisioning, fault management, and performance optimization across a multi-vendor network incredibly labor-intensive, error-prone, and slow. The shift towards Network Functions Virtualization (NFV) and Software-Defined Networking (SDN) in the 2010s exacerbated this problem, as networks became more software-based, dynamic, and composed of elements from numerous suppliers.

The motivation for FNIM was to provide a unified, standardized information framework that could serve as the foundation for the 3GPP's vision of automated, zero-touch network and service management. By defining a common model, it enables management systems from different vendors or responsible for different domains to understand and exchange information seamlessly. This solves the problem of semantic interoperability—ensuring that 'a virtual CPU core' or 'a network slice instance' means the same thing to all systems involved in its lifecycle.

Historically, it evolved from and integrates with other modeling efforts like the TM Forum's SID (Shared Information/Data) model and ETSI NFV's information models. Its development in 3GPP, starting in Release 11, was driven by the need for a telecom-specific model that could handle the unique requirements of 3GPP network technologies while enabling federation with broader IT management ecosystems. The FNIM is thus a key enabler for achieving the goals of 5G automation, network slicing (where each slice requires its own managed lifecycle), and efficient multi-domain service orchestration.