TORM: a hybrid multicast infrastructure for interactive distance learning

Multicast has been intensively studied over the past decade. Existing multicast services come in two-flavors, IP multicast and application level multicast, depending on which layer the service is implemented on. After investigating the pros and cons of both approaches, as well as the characteristics of interactive distance learning, we find it an imperative to combine these two multicast approaches. In this paper, we present TORM (totally ordered reliable multicast), a hybrid multicast infrastructure for interactive distance learning. Apart from blending IP multicast and application level multicast, TORM builds and maintains a tree structure upon which transport functionalities are implemented, including a differentiated error recovery service, flow and congestion control and global ordering