Multicast address management in the Internet: a study of the port blocking problem

Summary form only given. A critical issue in multicast communication is how to identify multicast groups. In the Internet, multicast addresses and port numbers distinguish multicast groups from each other. Multicast addresses are used within the network to route the multicast traffic to its destination hosts, whereas port numbers are used within the hosts to demultiplex multiple traffic among the host processes. For a process to join a multicast group, it must join both the address and the port number of the group. Within a host, port numbers can be considered as a limited resource. In general, if a port is already allocated to a process, it cannot be assigned to another one at the same time. This implies that a process may not be able to join a multicast group if the port number of that group is already occupied in the host by another process. This paper focuses on the port blocking problem. Mainly, the paper introduces models and techniques to measure multicast port blocking. First, it presents an experimental work in the study of the problem of port blocking. Second, it introduces two models of the problem: analytical model and simulation model. Results obtained from both the experimental work and the developed models show that port blocking plays an active role in multicast address management and the overall multicast communication.