摘要
参与协同的各个用户分别加入各局部组播组,通过计算节点的处理器能力、内存大小、带宽、接收队列状态及吞吐性能等性能参数,得出各节点的性能指标(即权重),从中选取最优者作为管理员,次优者作为管理员候选者,候选者在管理员失效的状况下成为管理员。管理员负责处理局部组播组的ACK报文及报文重传。并引入流量控制,使得大量的交互信息不会引起网络状况的严重恶化。并利用NS2对协议进行了仿真实现。
The participants,the members take part in a collaborative program,are divided into multiple local multicast groups. By calculating the parameters of every participants about ability of microprocessor,size of memory,bandwidth,receiving queue and throughput,the node in best performance becomes the manager of a group,the suboptimal one becomes the candidate for the manag-er which will be a new manager when the old one break down. In each of these groups,the processing of acknowledgement and re-transmission is assigned to its group manager. This paper add flow control into this protocol and ensure the network communication will not go down even if numerous collaborative information packets in the network.
出处
《微计算机信息》
北大核心
2008年第36期104-106,共3页
Control & Automation
关键词
可靠组播
流量控制
权重
reliable multicast
flow control
weight
