Peer Discovery for Disconnected SDN Nodes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In Software Defined Networks (SD-WANs), network management systems face challenges in identifying and reconnecting disconnected nodes when control plane connectivity is lost, as existing methods rely on control plane connections for health data monitoring and debugging.

Innovation Solution

A panic peer discovery process is implemented, where network management appliances detect lost connectivity and use peer edge network devices or controller appliances to receive health reports, allowing for debugging and reconnection of inaccessible nodes through alternative data plane connections.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If control plane connections are used for monitoring node health, then network management can effectively identify disconnected nodes, but connectivity loss prevents health data retrieval and node reconnection

Engineering Contradiction:
Improvenode health monitoring reliabilityVSAvoidhealth data loss when control plane disconnects
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The patent introduces a data plane connection as an intermediary communication channel between network management appliances and edge devices. When the control plane connection is lost, the data plane serves as a mediator to transmit health data and enable node identification and reconnection, thus resolving the information loss problem during control plane disconnection.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of repair

If control plane connections are used for node monitoring, then network administrators can debug and fix disconnected nodes, but connection loss eliminates this capability

Engineering Contradiction:
Improvenode debugging capabilityVSAvoidnode reconnection time
Core Design Contradiction:
Ease of repairVSLoss of time

Solution Approach 1:

The patent implements preliminary action by establishing data plane connections before control plane disconnection occurs. Health data collection and node identification processes are prepared in advance through the data plane, so when control plane connection is lost, the network management system can immediately utilize the pre-established data plane connection to retrieve health data and begin debugging without time loss.

Inventive Principle:
Principle #10Preliminary action

3Device complexity

If only control plane connections are used for communication, then protocol implementation is simplified, but alternative communication paths are unavailable during disconnection

Engineering Contradiction:
Improvecommunication protocol complexityVSAvoidcommunication path flexibility
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent applies universality by designing the data plane connection to serve multiple functions: it acts as a primary communication channel for data transmission and simultaneously serves as a backup channel for health data retrieval and node management when the control plane connection is lost. This multi-functionality provides communication path flexibility without significantly increasing protocol complexity.

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

Data Source

PatentEP4005187B1Peer discovery process for disconnected nodes in a software defined network
Publication Date: 2023.10.18 CISCO TECHNOLOGY INC
  • EP4005187B1 patent drawingFigure 1
  • EP4005187B1 patent drawingFigure 2
  • EP4005187B1 patent drawingFigure 3

AI summary

The present disclosure is directed to a peer node discovery process whereby a network management node can discover peers of inaccessible nodes that have lost connectivity to the network management node over the control plane and receive health report of the inaccessible nodes via the discovered peers. In one example, a method includes detecting a loss of connectivity to a network node; based on a type of the network node, performing one of a first process or a second process to obtain a health report of the network node, the first process and the second process including identification of at least one corresponding peer node from which the health report of the network node is to be received; and analyzing the health report to determine root cause of the loss of connectivity.