Link-Level NFV Architecture for Scalable VNF Traffic Handling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing NFV infrastructures face challenges in meeting scalability, throughput, latency, and reliability due to the use of commercial off-the-shelf hardware, leading to increased operational complexity, extra management costs, and inefficient handling of large traffic loads.

Innovation Solution

A link-level virtualization architecture that integrates compute and network resources, eliminating the need for external load balancers, provides a single user interface, and allows for dynamic resource allocation and scaling, enabling high availability and flexibility.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple VNFs are implemented in a computer server using COTS hardware, then cost is reduced, but performance and management efficiency deteriorate

Engineering Contradiction:
ImproveperformanceVSAvoidmanagement complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the VNF management into three distinct plane functions: Management Plane Function (MPF) for resource mapping and configuration, Control Plane Function (CPF) for protocol processing and state management, and Data Plane Function (DPF) for packet forwarding. This segmentation allows each component to be independently optimized and managed, resolving the contradiction between using multiple VNFs for cost reduction and maintaining performance while managing complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces virtualization link level architecture components as intermediaries between the COTS hardware and the VNFs. The MPF acts as an intermediary that maps logical resources to physical resources, while the CPF serves as an intermediary for protocol processing. These intermediaries enable efficient management of multiple VNFs on COTS hardware without sacrificing performance.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If a single VNF is implemented in a computer server, then hardware cost is reduced, but performance characteristics deteriorate compared to specialized hardware

Engineering Contradiction:
Improveperformance characteristicsVSAvoidhardware resources
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent implements a universal VNF architecture that can handle multiple network functions and protocols through the CPF, which processes various packet types (LACP, ARP, routing protocols, etc.). This multi-functional design allows a single VNF implementation to replace multiple specialized hardware devices, achieving comparable performance characteristics while utilizing standard COTS hardware resources.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent changes the operational parameters of standard COTS hardware through the virtualization layer. The MPF maps logical interfaces and resources, while the DPF aggregates packets and optimizes data flow. These parameter changes enable COTS hardware to achieve performance characteristics previously only available from specialized hardware, resolving the contradiction between using cheaper hardware and maintaining performance.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If external load balancers are used for managing multiple VNFs, then traffic distribution is improved, but operational complexity and costs increase

Engineering Contradiction:
Improvetraffic handling capabilityVSAvoidoperational complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent merges the load balancing functionality into the virtualization link level architecture itself. The MPF combines resource mapping with load distribution capabilities, and the CPF integrates protocol processing with traffic management. This merging eliminates the need for separate external load balancers, maintaining traffic handling capability while reducing operational complexity and costs.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The virtualization architecture provides self-service load balancing through the MPF and CPF components. The system automatically maps logical resources to physical resources and distributes traffic based on the virtual link configuration, without requiring external load balancer management. This self-service capability resolves the contradiction between improving traffic handling and reducing operational complexity.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS12531786B2Link-level network virtualization architecture for large-scale network function virtualization applications
Publication Date: 2026.01.20 FORTINET INC
  • US12531786B2 patent drawing
  • US12531786B2 patent drawing
  • US12531786B2 patent drawing

AI summary

A virtualized network function (VNF) of a network function virtualization (NFV) link level architecture includes a management plane function (MPF) to map a logical interface as a virtual interface in a control plane function (CPF) of the VNF. The MPF maps a logical resource of the NFV virtualization link level architecture as a shadow resource on one or more data plane functions (DPFs) of the VNF, wherein a virtual link is formed by the logical interface, the shadow resource, and the virtual interface. One of the DPFs receives a packet over the virtual link from a switch and classifies a type of the packet. In response to the packet type being a data packet, the DPF aggregates the packet with other data packets to form aggregated data packets and sends the aggregated data packets to the switch.