摘要
流量和拥塞控制的目的是限制网络中分组传输的平均时延和缓冲区溢出,并公平地处理各Session。基本的流量和拥塞控制的方法有两种。一是窗口式流量和拥塞控制,二是输入速率控制。主要阐述第一种方法,包括端到端的窗口流控、虚电路中的逐跳流控、ARPANET中的流控、SNA网中的流控以及PARIS网络中的流控等。在这些方式中,根据网络的拥塞情况,统一窗口拥塞控制、动态地调整窗口的大小。
The purpose of flow and congestion control is to restrict the average delay of packet transmission and the overflow of buffer area in the network, and solve every Session fairly. There are two methods about the basic flow and congestion control. One is window flow and congestion control; the other is the input rate control. The article focuses on the first method which include the end-to-end window flow control, the gradual flow control in virtual circuit, the unified congestion control, the flow control of ARPANET, the flow control in SNA net and the flow control in PARIS net and so on. According to the congestion situation of net, the volume of the widow can be modulated dynamically.
出处
《贵州大学学报(自然科学版)》
2009年第5期66-69,共4页
Journal of Guizhou University:Natural Sciences
基金
国家自然科学基金资助项目(60362001
69862001
F04241004)
关键词
窗口
虚电路
链路
公平性
window
virtual circuit
link circuit
fairness