基于排队论模型的网络拥塞率研究

Study of network congestion rate based on queuing theory model

分析了目前拥塞控制存在的问题,以排队论模型为基础对网络拥塞率进行了研究,从一个路由器的角度出发研究网络拥塞问题,提出了基本模型,给出了求解瞬时拥塞率的算法,并得到了稳态拥塞率的解析表达式。通过数值结果指出,路由器缓存的适当增加可以降低拥塞率,但过渡增加反而会增加拥塞率;增加路由器的CPU处理速度可以降低拥塞率。

The problem existing in congestion control is analyzed. The network congestion rate is researched based on queuing theory model, a basic model is proposed by studying the problem of network congestion from router, an algorithm for solving the instantaneous congestion rate is presented and the analytical formula of stable congestion rate is obtained. The numerical result shows that we can propefly decrease the congestion rate by increasing cache memory, however, overage cache memory may increase the congestion rate. Furthermore, by increasing the speed of CPU of the router can also decrease the congestion rate.