Network Congestion Control via Router Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing congestion control methods in IP internetworks either discard packets at routers, which does not consider intermediate hops, or use end-to-end protocols like TCP, which neglects router capacity, leading to inefficient data transmission.
Innovation Solution
Implementing a congestion control method that establishes packet queues with rate values and intervals for each network, allowing packets to be sent based on the maximum capacity of routers, ensuring efficient end-to-end transmission by managing packet rates and intervals dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If routers discard packets to control congestion, then packet loss is reduced, but intermediate hop capacity is not considered
Solution Approach 1:
The patent implements feedback mechanisms where routers send congestion notifications back to source nodes based on actual queue depths and capacity utilization at intermediate hops. This allows source nodes to adjust transmission rates dynamically, considering the capacity constraints of intermediate routers rather than just end-to-end connectivity.
Solution Approach 2:
The patent introduces intermediate routers as active mediators in the congestion control process. These routers monitor their own capacity utilization and actively participate in regulating traffic flow by sending feedback signals, rather than merely discarding packets when overwhelmed.
2Ease of operation
If end-to-end protocols like TCP are used for congestion control, then source and destination nodes can coordinate, but router capacity is neglected
Solution Approach 1:
The patent segments the end-to-end congestion control problem into multiple localized control regions at each intermediate router. Each router independently manages its own queue and capacity constraints, making smaller, more manageable control decisions rather than relying on a single end-to-end protocol to handle the entire path.
Solution Approach 2:
The patent adds a new dimension to congestion control by introducing intermediate router participation between the source and destination nodes. This creates a multi-layered control architecture where both end-to-end coordination and intermediate hop capacity management operate simultaneously, rather than relying solely on end-to-end protocols.
3Productivity
If packet rates are increased to maximize throughput, then network utilization improves, but congestive collapse occurs
Solution Approach 1:
The patent implements preliminary congestion avoidance actions by monitoring queue depths and capacity utilization before congestive collapse occurs. Routers send early warning signals to source nodes when queues approach capacity thresholds, allowing proactive rate adjustment rather than reactive packet discarding after collapse has begun.
Solution Approach 2:
The patent introduces dynamic packet rate adjustment mechanisms that continuously adapt transmission rates based on real-time network conditions. Rather than maintaining fixed high rates that risk collapse, the system dynamically modulates rates according to intermediate router capacity constraints, balancing throughput maximization with stability maintenance.
Data Source
AI summary
Systems and methods of optimizing packet flow in a network are disclosed. An internetwork includes plural networks, each network having plural non-router nodes and at least one router node. The non-router nodes of a first network can be configured to perform congestion control. The non-router node establishes a packet rate value for a packet queue associated with a second network of the internetwork. A total number of packets not greater than the packet rate value are removed from the packet queue. The removed packets are sent to a node in the second network.