Novel multicast address allocation approach based on interest constrains

There exists a sharp conflict between the large amount of multicast addresses required in distributed virtual environment and the multicast address finiteness. In this paper, a novel multicast address allocation approach based on the interest constrains is proposed, which operates the allocation process with a seeding model. The initiation and run-time process of seeding model carry out the static and dynamic multicast address allocations individually. This paper combines 2-dimension lattice and PR quadtree structure to design the spatial data structure of seeding model and presents the adaptive growing and pruning algorithms for dynamic allocation, together with the searching algorithm for rapid searching the multicast address for a location. Algorithm analysis and performance experiments show that this approach has high efficiency and reach the requirements for multicast address allocation and search in large-scale distributed virtual environment. In the end a brief introduction to the implementation of this approach in BH-RTI, a runtime infrastructure, for distributed simulation is given.