Predictive Fast Reroute for Network Resilience
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional network recovery mechanisms are reactive and agnostic to the quality of experience (QoE) of applications, leading to inefficient traffic rerouting after link or node failures, with recovery times often measured in milliseconds but lacking proactive predictive capabilities.
Innovation Solution
A router in a network receives a prediction model from a supervisor to anticipate failures along a primary path, enabling proactive fast rerouting of traffic to a backup path before the failure occurs, leveraging real-time prediction and machine learning to assess and improve the quality of service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reactive recovery mechanisms are used, then network recovery is achieved after failure detection, but recovery time is increased and QoE is not optimized
Solution Approach 1:
The system performs preliminary actions by predicting potential network failures before they occur and proactively rerouting traffic in advance. The prediction model analyzes historical data, network topology, and real-time metrics to identify at-risk links and nodes, enabling the router to switch to backup paths before actual failures happen, thus reducing recovery time and improving reliability
2Speed
If traditional FRR is deployed, then fast rerouting is achieved after failure detection, but the system remains reactive and cannot prevent failures
Solution Approach 1:
The system implements feedback mechanisms by continuously monitoring network metrics (link utilization, packet loss, jitter) and using this information to train prediction models. The models learn from historical failure patterns and real-time data to predict future failures, enabling proactive rerouting that prevents disruptions before they affect users
3Productivity
If reactive rerouting is used, then traffic is routed after failure occurs, but quality of experience for applications is not optimized
Solution Approach 1:
The system changes parameters by transitioning from reactive to predictive operation modes. The prediction model processes multiple parameters including historical failure data, current network state, and application requirements to determine optimal rerouting decisions. This enables the system to consider QoE metrics and application-specific needs when making rerouting decisions, rather than using generic failure response
Data Source
AI summary
In one embodiment, a router in a network reports, a supervisor, capabilities of the router to support fast reroute. The router receives a prediction model from the supervisor that is able to predict failures along a path in the network. The router predicts, using the prediction model, a failure along a primary path in the network that is currently being used by the router to send traffic. The router performs, in advance of the failure predicted by the router, a fast reroute of at least a portion of the traffic from the primary path to a backup path in the network.