Network Device Congestion Flow Identification via Port Utilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current congestion flow identification methods in network devices suffer from low accuracy due to random selection of data packets and reliance on queue length thresholds, leading to inefficient handling of head-of-line blocking.
Innovation Solution
A network device determines the target output port utilization by querying the forwarding information base and isolates congestion flows to a dedicated queue when the utilization exceeds a threshold, improving identification accuracy by monitoring port usage and flow control status.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If random selection of data packets is used for congestion flow identification, then the identification process is simple, but the congestion flow identification accuracy is low
Solution Approach 1:
The patent changes the identification parameters from random packet selection to monitoring queue length changes over time. By tracking whether the queue length increases when a data packet is transmitted, the system identifies congestion flows with high accuracy while maintaining implementation simplicity through straightforward queue length monitoring.
2Ease of operation
If queue length threshold is used to determine congestion, then the determination method is simple, but the congestion flow identification accuracy is low
Solution Approach 1:
The patent introduces feedback by continuously monitoring queue length changes before and after data packet transmission. The system uses the feedback from queue length variations to dynamically identify congestion flows, achieving high accuracy while keeping the operation simple through automated threshold-based detection of queue length changes.
3Ease of operation
If head of line blocking is not addressed, then the queue operates in first in first out manner, but data packets that can be forwarded are blocked
Solution Approach 1:
The patent segments the queue into congestion flow queues and non-congestion flow queues based on real-time identification. By dividing the single FIFO queue into multiple specialized queues, the system eliminates head of line blocking for non-congestion flows while maintaining simple FIFO operation within each segment, thereby improving overall forwarding efficiency.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of this application disclose a congestion flow identification method and a network device, to improve congestion flow identification accuracy. The method in the embodiments of this application includes: obtaining, by a network device, a queue length of a non-congestion flow queue, where the non-congestion flow queue includes a data packet or description information of the data packet; determining, by the network device, a target output port of a target data packet when the length of the non-congestion flow queue is greater than or equal to a first threshold, where the target data packet is a data packet waiting to enter the non-congestion flow queue or a next data packet waiting to be output from the non-congestion flow queue; and when utilization of the target output port is greater than or equal to a second threshold, determining, by the network device, that a flow corresponding to the target data packet is a congestion flow. The embodiments of this application further provide a network device, to improve congestion flow identification accuracy.