A Sketch-Based Fine-Grained Proportional Integral Queue Management Method

The phenomenon “bufferbloat” occurs when the buffers of network intermediary nodes fill up, causing long queuing delays. This has a significant negative impact on quality service applications, particularly those that are sensitive to time delay. Many active queue management (AQM) algorithms have been proposed overcome this problem. Those AQMs attempt maintain minimal delays and good throughput by purposefully dropping packets at nodes. However, existing AQM mostly drop randomly based certain metric such as length or delay, which fails achieve fine-grained differentiation data streams. In paper, we propose sketch-based proportional integral algorithm S-PIE, uses an additional measurement structure Sketch for packet frequency share judgment PIE between streams adjust policy differentiated drop. Experimental results NS-3 simulation platform show S-PIE achieves lower average RTT higher fairness than PIE, RED, CoDel while maintaining similar performance, availability stability, improving service.