Token Scheduler Aggregation for Fair Traffic Shaping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current traffic shaping methods, such as WFQ, face challenges in evenly distributing tokens among data streams, leading to inefficient bandwidth utilization, especially when multiple data streams have similar small rates, resulting in token concentration in idle streams and inadequate token allocation.
Innovation Solution
A method and apparatus that configure weights for each data stream, aggregate them into a total data stream token scheduler, and distribute tokens evenly through a top token scheduler, ensuring fair scheduling and improved bandwidth utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If WFQ method is used to add tokens for multiple data streams with small rates, then token distribution follows weight proportion, but tokens are intensively added into idle stream and no data stream acquires token during a period of time
Solution Approach 1:
The patent segments the token distribution process into multiple phases: first distributing tokens to data streams with larger rates, then distributing remaining tokens to data streams with smaller rates. This segmentation prevents token concentration in idle streams and ensures all data streams can acquire tokens periodically, thereby improving bandwidth utilization while maintaining operational simplicity.
2Reliability
If WFQ method allocates bandwidth proportionally according to weight, then allocation is fair, but when sum of shaping rates exceeds maximum weight, no idle stream exists and data streams cannot acquire deserved tokens
Solution Approach 1:
The patent implements a dynamic token distribution mechanism that adjusts distribution strategy based on real-time conditions. When the sum of shaping rates exceeds maximum weight, the system dynamically switches to a two-phase distribution approach: first satisfying data streams with larger rates, then distributing remaining tokens to data streams with smaller rates. This dynamic adjustment maintains fairness while enabling full bandwidth utilization.
3Adaptability or versatility
If tokens are distributed according to weight proportion in WFQ, then allocation follows configured weights, but number of tokens acquired by each data stream cannot reach the number deserved by configured weight
Solution Approach 1:
The patent introduces a new dimension to token distribution by implementing a two-phase approach: first dimension distributes tokens to data streams with larger rates, second dimension distributes remaining tokens to data streams with smaller rates. This dimensional change allows the system to honor configured weights while ensuring all data streams acquire their deserved tokens, thereby improving token acquisition efficiency without sacrificing allocation flexibility.
Data Source
AI summary
A method and an apparatus for adding tokens in traffic shaping are provided. According to the method, a weight of each data stream is configured, and each data stream is added to corresponding token schedulers based on the weight of the each data stream. Based on a weight of each token scheduler, each token scheduler is sequentially added to corresponding token scheduler groups, and all the token scheduler groups are aggregated into a token scheduler of a total data stream. The token scheduler of the total data stream is added to a top token scheduler. Based on a preset cycle, a token is assigned to the token scheduler of the total data stream in the top token scheduler, and the assigned token is delivered to each token scheduler group or to a respective data stream.