Switching Network Back-Pressure Priority Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In high-capacity packet switching communication systems, data streams of different service levels are not prioritized effectively, leading to congestion where high-priority data is not transmitted preferentially, occupying buffer space and reducing throughput.
Innovation Solution
A data transmission method that determines a back-pressure priority based on current queue length and a mapping relationship, generating back-pressure information to inhibit line cards from sending data with priorities below or equal to the determined priority, ensuring high-priority data is transmitted efficiently by prioritizing buffer space allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a queue back-pressure mechanism is used to stop data transmission when congestion occurs, then buffer overflow is prevented, but high-priority data streams cannot be transmitted preferentially and throughput of high-priority data is reduced
Solution Approach 1:
The patent applies local quality by differentiating the back-pressure mechanism based on data priority levels. Instead of a uniform back-pressure approach, the system generates different back-pressure priorities corresponding to different queue lengths, allowing high-priority data to be transmitted preferentially even when congestion occurs. The switching network card determines back-pressure priority based on current queue length and mapping relationships, creating localized quality differences in the back-pressure response.
2Quantity of substance
If buffer space is allocated to low-priority data, then more data can be stored, but buffer space for high-priority data is reduced and high-priority throughput is limited
Solution Approach 1:
The patent implements dynamic buffer space allocation through dynamic back-pressure priority assignment. The back-pressure priority is not fixed but is determined dynamically based on the current queue length and pre-configured mapping relationships. When the queue length increases, the back-pressure priority increases, automatically adjusting the buffer allocation behavior to favor high-priority data transmission under congestion conditions while maintaining total buffer capacity.
3Reliability
If all data streams are stopped simultaneously when congestion occurs, then buffer overflow is prevented, but service level differentiation is lost and high-priority data cannot be prioritized
Solution Approach 1:
The patent applies parameter changes by modifying the back-pressure priority parameter based on queue length conditions. The system uses pre-configured mapping relationships between queue length ranges and back-pressure priorities to dynamically adjust the back-pressure parameter. This allows the system to maintain buffer overflow prevention while simultaneously preserving service level differentiation through parameter variation based on congestion severity.
Data Source
Figure 1~2
Figure 3
Figure 4~6
AI summary
A data transmission method is provided, which includes: obtaining a current queue length of a queue corresponding to an output port; when the current queue length meets a back-pressure requirement, determining a back-pressure priority corresponding to the current queue length according to the current queue length and a mapping relationship between a preset queue length and the back-pressure priority, and generating back-pressure information, where the back-pressure information inhibits a line card from sending data with a data priority less than or equal to the back-pressure priority to the output port; and sending the back-pressure information to a line card. Through the method, the back-pressure information is generated according to the mapping relationship between the current queue length, the preset queue length and the back-pressure priority to inhibit all the line cards from sending the data with a data priority less than or equal to the back-pressure priority, which alleviates a problem that a great deal of data with a low priority occupies a buffer of a switching network over a long period of time, so that the buffer reserved for the data with a high priority is increased, thereby increasing the high throughput of data with a high priority in the switching network.