Network Switch Adaptive Routing via Aggregated Congestion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional adaptive routing technologies in large network infrastructures, such as data centers, often rely on static topology information and aggregate congestion values, which are inadequate for dynamic routing decisions, leading to inefficiencies in network traffic management.
Innovation Solution
A network switch that aggregates congestion information from multiple output buffer queues to determine the maximum occupancy count for each output port, calculates local and remote congestion values, and selects the output port with the lowest total congestion value for routing network packets, thereby optimizing traffic flow and avoiding saturated paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If static topology information is used to determine network topology, then the network infrastructure can be managed with simpler methods, but adaptive routing technologies cannot effectively utilize this information for dynamic routing decisions
Solution Approach 1:
The patent transitions from static topology information to dynamic topology information by having computing nodes continuously advertise their current topology state to neighbors. This allows the system to adapt to changing network conditions in real-time, enabling adaptive routing decisions based on current network state rather than predetermined static configurations.
Solution Approach 2:
The patent implements feedback mechanisms where computing nodes exchange topology information with their neighbors, allowing each node to be aware of the current topological state. This feedback loop enables dynamic routing decisions by continuously updating routing tables based on real-time network conditions, resolving the contradiction between static management simplicity and dynamic adaptability.
2Device complexity
If aggregate congestion values are used for routing decisions, then the routing process is simplified, but the ability to make precise routing decisions based on actual buffer queue saturation is lost
Solution Approach 1:
The patent segments the aggregate congestion value into individual buffer queue-level congestion measurements. Instead of using a single aggregated congestion value, the system measures and tracks the saturation state of each individual buffer queue, allowing for more precise routing decisions that account for the specific state of each output port's buffer queue.
Solution Approach 2:
The patent applies local quality by measuring congestion at the individual buffer queue level rather than using a global aggregate value. Each buffer queue's saturation state is independently measured and used for routing decisions, allowing the system to make precise local routing choices based on the actual congestion state of each specific output port's buffer queue.
3Stability of the object's composition
If traditional routing methods are used in large network infrastructures, then the network can maintain stable operation, but network efficiency and traffic management performance deteriorate
Solution Approach 1:
The patent introduces dynamic routing decisions based on real-time topology and congestion information while maintaining stable network operation. The system continuously updates routing tables based on current network conditions, allowing the network to adapt to changing conditions without sacrificing stability, thereby improving overall network efficiency in large infrastructures.
Data Source
AI summary
Technologies for adaptive routing based on aggregated congestion information include a network switch that includes a plurality of output ports. The network switch is configured to determine a maximum local occupancy count for each output port based on a maximum local occupancy count of output buffer queues of each output port, a local congestion value based on the maximum local occupancy count, and a remote congestion value for a corresponding remote input buffer queue of a remote computing device communicatively coupled to a corresponding output port. The network switch is further configured to determine, for each output port, a total congestion value as a function of the local congestion value and the remote congestion value and enqueue the network packet into one of the output buffer queues of one of the output ports based on the total congestion values of the output ports. Other embodiments are described herein.