An overlay network construction method using swarm intelligence for an application-level multicast service

In an overlay network, each peer participating a session forwards messages requested by other peers to embody a multicast service at an application layer. Therefore, the topology of an overlay network influences on the quality of service provided to each peer as well as on the number of participants that the overlay network can support. As the capacity of a peer increases, it should support more children peers and be located closer to a multicast source to increase the resource efficiency of an overlay network. However, since a peer uses its local information when it chooses a parent peer that it requests data from, it is unlikely that the resulting topology of an overlay network is optimal. To resolve this problem, we propose an overlay construction method using the swarm intelligence of peers. In our method, a peer maintains the states information of a small set of peers, and a new peer uses the swarm intelligence provided by a set of peers to search for a desirable parent peer in terms of the distance from a multicast source and the capacity of a peer. The simulation results show that a peer support more children peers as the capacity of a peer increases. Therefore, it helps to prevent a high capacity peer from being limited by a low capacity parent peer. In addition, the proposed method is superior to the other methods in terms of the maximum distance from a multicast source to a peer.