Self-Healing Network Loop Detection via Steady State Flow Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unwanted network loops in communications networks cause network broadcast storms, leading to bandwidth deterioration and wasted CPU utilization due to duplicate packets.
Innovation Solution
A method that collects steady state network traffic flow data, samples real-time data flows, and compares them to detect loops, using loop pointers to identify and remove unwanted loops, enabling self-healing and self-diagnosing networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If network loops are allowed for network flexibility and redundancy, then network adaptability is improved, but network stability deteriorates due to broadcast storms and duplicate packets
Solution Approach 1:
The system performs preliminary actions by collecting steady state network traffic flow data before loops occur, establishing a baseline of normal unidirectional flow patterns. This pre-established reference data enables the system to detect loops by comparing real-time traffic against the stored steady state, allowing preventive detection before broadcast storms damage network stability
Solution Approach 2:
The system implements feedback by continuously monitoring network traffic flows and comparing them against stored steady state data. When bidirectional flows are detected that contradict the established unidirectional steady state patterns, the system generates alerts and can trigger automated responses such as blocking problematic ports, creating a closed-loop control system that maintains network stability while allowing flexibility
2Reliability
If network monitoring is implemented to detect loops, then network stability is improved, but device complexity increases due to additional monitoring infrastructure
Solution Approach 1:
The system applies self-service by utilizing existing network traffic data that is already flowing through the network for monitoring purposes. Instead of requiring separate probe devices or additional infrastructure, the system leverages the network's own traffic flows as the monitoring medium, analyzing packet metadata and flow patterns that naturally occur during normal network operation
Solution Approach 2:
The system demonstrates multi-functionality by using a single monitoring mechanism that simultaneously performs multiple tasks: detecting loops, identifying broadcast storms, locating duplicate packets, and providing network performance analysis. This universal approach consolidates what would traditionally require multiple specialized monitoring tools into one integrated solution
3Measurement precision
If real-time loop detection is performed by comparing sampled data with steady state data, then loop detection precision is improved, but loss of time increases due to continuous data collection and comparison
Solution Approach 1:
The system performs preliminary action by pre-collecting and storing steady state network traffic flow data during normal operation before loops occur. This baseline data includes source-destination pairs, bandwidth utilization patterns, and flow directions. When loops are suspected, the system compares real-time sampled data against this pre-established reference, enabling rapid detection without requiring extensive real-time analysis from scratch
Solution Approach 2:
The system applies partial action by sampling network traffic at strategic intervals rather than continuously analyzing every packet. The monitoring system collects representative samples of network flows and compares these against steady state data, achieving sufficient detection precision without the computational overhead of exhaustive real-time analysis of all network traffic
Data Source
AI summary
Methods, apparatus, and computer-readable media for detecting and removing unwanted loops in communications networks having host processors, nodes, and segments connecting the nodes are disclosed. In one embodiment, a set of steady state data regarding direction of a network traffic flow is collected and stored. Then, a newly sampled flow is compared with the stored steady state data to detect flows whose paths contradict their steady state direction, and a network loop is identified when a segment carries such contradictory flows.