Supporting scalable peer to peer virtual environments using frontier sets

We present a scalable implementation of a network partitioning scheme that we have called frontier sets. Frontier sets build on the notion of a potentially visible set (PVS). In a PVS, a world is sub-divided into cells and for each cell all the other cells that can be seen are computed. In contrast, a frontier set considers pairs of cells, A and B. For each pair, it lists two sets of cells, F/sub BA/ and F/sub BA/. By definition, from no cell in F/sub BA/ is any cell in F/sub BA/ visible and vice-versa. Our initial use of frontier sets has been to enable scalability in distributed networking. In this paper we build on previous work by showing how to avoid pre-computing frontier sets. Our previous algorithm, required O(N/sup 3/) space in the number of cells, to store pre-computed frontier sets. Our new algorithm pre-computes an enhanced potentially visible set that requires only O(N/sup 2/) space and then computes frontiers only as needed. Network simulations using code based on the Quake II engine show that frontiers have significant promise and may allow a new class of scalable peer-to-peer game infrastructures to emerge.