Network Congestion Management via Backwards Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network traffic congestion leads to increased delays and potential system failures due to oversubscription of network links, where existing systems often rely on packet dropping or rate adjustments that are inefficient and resource-intensive.
Innovation Solution
A system that monitors congestion point queues, generates consolidated feedback parameters, and sends feedback messages to adjust network traffic rates, using techniques like backwards congestion notification (BCN) to mitigate congestion by shifting traffic from congested core areas to less aggregated edge areas with available resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If device level algorithms are used to handle congestion by dropping packets, then transient congestion is alleviated, but network resource waste increases and reliability deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where network devices monitor queue lengths and send feedback signals to source devices. When the queue length exceeds a threshold, feedback is sent to reduce the transmission rate, enabling dynamic adaptation without packet dropping.
Solution Approach 2:
The patent uses preliminary actions by setting equilibrium queue levels and offsets before congestion occurs. The system proactively adjusts transmission rates based on predicted queue states, preventing congestion before it leads to packet loss.
2Reliability
If transport level algorithms are used to adjust transmission rates, then sustained congestion is alleviated, but system complexity increases
Solution Approach 1:
The patent changes parameters by using equilibrium queue levels and offsets as simplified metrics for congestion control. Instead of complex multi-parameter optimization, the system adjusts transmission rates based on changes in these simplified queue parameters.
3Measurement precision
If detailed queue state information is transmitted for congestion control, then congestion management precision is improved, but bandwidth consumption increases
Solution Approach 1:
The patent extracts only the essential congestion information (equilibrium queue level and offset) from the complete queue state. This selective extraction provides sufficient congestion management precision while minimizing the bandwidth required for feedback transmission.
Data Source
AI summary
A method and system to manage network congestion are provided. In one example embodiment, the method comprises receiving an indication of a rate increase request at a reaction point computer system, determining information indicative of a frequency of rate decrease requests during a period of time, and initiating a rate increase signal utilizing the information indicative of the frequency of rate decrease requests during the period of time.


