二进制流量控制算法的性能分析

自适应比特(available bit rate,简称ABR)流量控制是ATM网络中拥塞控制和流量管理的一种有效手段.在大规模的高速网络中,简洁、有效的实现算法对优化交换机的性能是至关重要的.二进制流量控制策略以其实现上的简洁性吸引了广泛的注意,但标准EFCI算法作用下队列和速率的振荡却给性能造成了负面影响,使人们对二进制流量控制机制产生了疑虑,不得不放弃二进制算法的简洁性,转而研究相对复杂但有效的显式速率反馈算法.在已建立模型的基础上,用非线性控制理论中描述函数的分析方法系统地评价和分析了标准EFCI算法的性能.得到的结论是:依赖于直觉的启发式非线性控制算法诱发的自激振荡是造成队列和速率振荡的本质原因,而非二进制流量控制机制本身固有的属性,仿真试验的结果证实了理论分析的结论.最后,提出了一种通过修改交换机参数配置策略来优化现已有EFCI交换机性能的方案.

Performance Analysis of the Binary Flow Control Algorithm

ABR (available bit rate) flow control is an effective measure in ATM network congestion control and traffic management. In large scale and high-speed network, the simplicity of the algorithm is crucial to optimize the switch performance. Though the simplicity of binary flow control is very attractive, the queue length and allowed cell rate (ACR) controlled by the standard EFCI algorithm oscillate with great amplitude, which has negative impact on the performance, so its applicability is doubted, and then relatively complex but effective explicit rate feedback algorithms are introduced and explored. In this study, based on the existed flow control model, the performance of standard EFCI algorithm is evaluated and analyzed with the describing function approach in nonlinear control theory, concluding that queue and cell rate self-oscillations are caused by the inappropriate nonlinear control law originated from intuition, but not intrinsic attribute of the binary flow control mechanism. The simulation experimental results are done to validate this analysis and conclusion. Finally, a parameter settings scheme is put forward to optimize the existed EFCI switch.