Weighted Random Early Back-to-Sender Signaling for Network Congestion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing congestion signaling in computing networks results in additional bandwidth overhead due to frequent end-to-end notifications, and existing congestion control mechanisms are inefficient in managing network paths and reducing latency.
Innovation Solution
Implementing Weighted Random Early Back-to-Sender (WREB) notifications that vary frequency based on congestion levels, using entropy values for path selection, and optimizing congestion control mechanisms to reduce overhead and improve bandwidth utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If frequent end-to-end congestion notification messages are sent, then congestion control reliability is improved, but network bandwidth overhead increases
Solution Approach 1:
The patent segments the congestion notification function by introducing intermediate network devices (switches, routers) that generate and forward congestion notifications independently. This divides the end-to-end notification path into multiple segments, allowing congestion signals to be propagated without requiring full end-to-end message exchanges, thereby reducing bandwidth overhead while maintaining control reliability.
Solution Approach 2:
The patent introduces intermediary network devices that act as mediators in the congestion notification process. These intermediaries generate congestion notifications based on local congestion detection and forward them to sending devices, eliminating the need for frequent end-to-end acknowledgment messages and reducing overall bandwidth consumption while maintaining effective congestion control.
2Stability of the object's composition
If traditional congestion control mechanisms are used, then network stability is maintained, but latency increases
Solution Approach 1:
The patent implements preliminary congestion notification by having network devices proactively generate and forward congestion notifications before packets experience actual delays. This early warning mechanism allows sending devices to adjust transmission rates in advance, preventing congestion-related latency while maintaining network stability through gradual rate adjustments.
Solution Approach 2:
The patent implements a feedback mechanism where network devices continuously monitor congestion conditions and send notifications back to sending devices. This real-time feedback loop enables dynamic adjustment of transmission rates, reducing latency by responding to congestion conditions faster than traditional mechanisms while maintaining stability through controlled rate adjustments.
3Productivity
If entropy values are saved and reused for path selection, then network path optimization is improved, but storage requirements increase
Solution Approach 1:
The patent applies local quality by maintaining entropy value storage structures at specific network devices (senders and intermediaries) rather than universally. Each device maintains only the entropy values relevant to its local path selection needs, optimizing path selection performance while minimizing total storage requirements across the network through localized, context-specific entropy caching.
Data Source
AI summary
It is determined that a computing node is contributing to a network congestion event. A congestion notification message is generated. A timing profile is determined for sending the congestion notification message based on the level of the network congestion event. Based on the timing profile, the congestion notification message is forwarded to the computing node determined to be contributing to the network congestion event.