Analyzing stream building processes in P2P streaming systems

In this paper, we analyze the stream building processes in P2P streaming systems. A stream building process (SBP) is defined as the process that a peer tries to build enough download streams. We argue that a SBP can be roughly seen as a stochastic process composed by random events of stream test, and the time cost of SBP statistically depends on the sequence of requested peers. By modeling the probability of successful request and peers' behaviors of joining and leaving, we construct a simulation system of P2P streaming with two alternative strategies. We trace the distribution of the time costs of SBPs in the system, and show our strategy can reduce at least 23% time cost of streaming building processes.