ALMI: An Application Level Multicast Infrastructure

Abstract

The IP multicast model allows scalable and efficient multi-party communication, particularly for groups of large size. However, deployment of IP multicast requires substantial infrastructure modifications and is hampered by a host of unresolved open problems such as reliability, flow and congestion control, security and access control. Motivated by these problems, we have designed and implemented ALMI, an application level group communication middleware, which does not rely on network infrastructure support and thus, allows accelerated deployment and simplified configuration at the cost of a relatively small increase in traffic load. ALMI is tailored toward support of multicast groups of relatively small size (several 10s of members) with many to many semantics. Participants of a multicast session are connected via a virtual multicast tree, i.e., a tree that consists of unicast connections between end hosts. The tree is formed as a minimum spanning tree (MST) where the cost of each link is an application-specific performance metric. The shift of multicast to end systems introduces certain performance penalties, such as duplicate packets on physical links and larger end-to-end delay than IP multicast. Using simulation, we show that the performance tradeoff is quite small and that ALMI multicast trees approach the efficiency of IP multicast trees. We have also implemented ALMI as a Java based middleware package and performed experiments over the Internet. The experimental results show that ALMI is able to cope with network dynamics and keep the multicast tree efficient.