Congestion Notification Control Loop Parameter Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network congestion leads to increased delays and potential application failures due to overloaded network pathways, where existing congestion management mechanisms like Backward Congestion Notification (BCN) struggle to effectively adjust traffic rates across the network.
Innovation Solution
A system and method to adjust Congestion Notification (CN) control loop parameters by monitoring the congestion point queue, generating feedback messages with responsiveness data, and applying rate limiting at reaction points to mitigate congestion, shifting traffic load from network cores to edges where resources are more available.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If rate-based closed loop feedback control mechanism is implemented with core detector integrated in hardware, then congestion detection capability is improved, but device complexity increases
Solution Approach 1:
The congestion detection system uses self-service by having the network infrastructure automatically monitor and report its own congestion state through BCN messages, eliminating the need for external monitoring devices and reducing overall system complexity while maintaining high detection capability
Solution Approach 2:
The patent implements feedback by having core detectors send BCN messages back to reaction points when congestion is detected, creating a closed-loop control system that automatically adjusts traffic rates based on real-time congestion measurements, resolving the contradiction between detection precision and system complexity
2Measurement precision
If BCN messages include congestion change information and instantaneous measures, then control precision is improved, but loss of information increases due to message overhead
Solution Approach 1:
The patent extracts only the essential congestion parameters needed for control decisions into BCN messages, separating critical information (congestion state, queue length) from detailed raw data, thereby achieving precise control while minimizing message overhead and information loss
Solution Approach 2:
The system changes parameters by encoding congestion information in standardized BCN message formats with specific fields for congestion state and queue metrics, optimizing the balance between control precision and message size to reduce overhead while maintaining necessary control accuracy
3Reliability
If rate limiters shape individual flows at reaction points, then congestion mitigation effectiveness is improved, but device complexity increases
Solution Approach 1:
The patent applies segmentation by implementing rate limiters that shape individual flows separately at reaction points, dividing the congested traffic into manageable per-flow controls rather than treating all traffic uniformly, which improves mitigation effectiveness while keeping each rate limiter component relatively simple
Solution Approach 2:
Reaction points perform self-service by automatically adjusting their own rate limiting parameters based on received BCN messages, eliminating the need for centralized control logic and reducing overall system complexity while maintaining effective congestion mitigation through distributed autonomous control
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and system to adjust Congestion Notification control loop parameters at a congestion point are provided. The system comprises a monitor to sample a state of a congestion point the congestion point being to receive messages from a reaction point; a history generator to generate an updated reaction to congestion history by consolidating the state of the congestion point with a current reaction to congestion history, the current reaction to congestion history being associated with a current feedback message; a message generator to generate an updated feedback message based on the updated reaction to congestion history; a message update module to replace the current feedback message with the updated feedback message; and a communications module to communicate the updated feedback message to the reaction point.