Application-Layer Multicast (ALM)
When the same data is needed at many locations, the sender does not have to supply every receiver directly. A receiver can forward the data to others. Application-layer multicast (ALM) builds such a distribution tree between applications, using ordinary IP connections rather than multicast functions in the routers.
The shape of that tree matters. A direct connection can give a receiver a short path, while forwarding through another node may use the available capacity differently. FILA investigates how to choose these connections so that data reaches the group efficiently without taking unnecessarily long routes.
© FILA
The application distributes data through a tree: some nodes receive and forward it, while others form the endpoints.
Each branch is a choice about where data travels next. FILA uses measurements of round-trip time and available bandwidth to assess possible connections. These quantities describe different aspects of a path: a link with favourable capacity is not necessarily the link with the shortest delay.
The 2025 study combines these objectives when constructing a tree and adds a tunable limit on source-to-receiver path length. Evaluated with RMDT in a five-region cloud testbed, the method lets the latency requirement influence the tree instead of optimising every connection for a single metric.
The tree is built from measurements, so their freshness and the resources of forwarding nodes matter. As traffic changes, a useful topology may change too. This connects topology selection with transport and measurement: the question is how the whole receiver group behaves, rather than how one attractive link performs.
Selected publications
- Kirill Karpov, Dmitry Kachan, Eduard Siemens (2023). Enhancing Application Layer Multicast Using Multi-objective Optimization Techniques. Information and Communication Technologies and Sustainable Development, Lecture Notes in Networks and Systems 809, 42–53 (Springer) DOI: 10.1007/978-3-031-46880-3_3.
- Kirill Karpov, Dmitry Kachan, Veronika Kirova, Ali Ghermezian, Daria Koshutina, Eduard Siemens (2025). Tunable Multi-Objective Tree Synthesis for Application-Layer Multicast. Proceedings of the International Conference on Applied Innovations in IT (ICAIIT), 13(5), 69–74 DOI: 10.25673/122809.
