拥塞控制模型在XMSF组播中的研究与性能验证

由于可扩展建模与仿真框架（XMSF）提出的整套方法是建立在与Internet相结合的性能基础之上的，因而底层网络提供的性能在XMSF中起着十分重要的作用。为了满足XMSF网络服务需求，需采用叠加组播技术提供多对多的通信服务，同时应使用拥塞控制机制以保证网络服务质量（Qos）。该文首先介绍了叠加组播的概念，同时和传统的IP组播进行了比较，然后给出了一种组播拥塞控制模型，并在MAX—PLUS代数基础上采用线性循环法，分析了该模型吞吐量的计算方法，最后通过仿真验证了该模型的可行性和灵活性。

The Technology of Congestion Control and Its Implementation in XMSF Multicast

Because the whole set of measures brought out by XMSF is based on the performance associated with Internet, so the performance provided by underlying networks plays a very important role in XMSF. In order to meet network service requirements, many-to-many multicast should be provided in a way of overlay multicast. And a congestion control mechanism ought to be used to ensure the Network Quality of Service (Qos). This paper introduces the concept of overlay multicast, and makes a comparison with traditional IP multicast, then presents a multicast congestion control model, and uses a linear recurrence method based on the max-plus algebra to calculate the throughput. Finally, the results of the simulation prove the feasibility and reliability of the model.