A Game-Based Mechanism for Avoiding Routing Hotspot in P2P Streaming Distribution

Today´s P2P streaming application periodically suffers from routing hotspots, a.k.a., flash crowds. A routing hotspot is typically created by an unanticipated new event that triggers an unanticipated surge of users which request streaming from some particular peers, temporarily overwhelming the peer´s delivery capabilities. In this paper, we consider how to avoid the routing hotspot prior to a congestion event. Assume that the price for usage of each node is proportional to the congestion level at the node, we propose an incentive compatible pricing mechanism. It is shown that the users can enforce any operating point it deems efficient and avoid hotspot by offering the capacity of the various nodes at weight prices. The incentive compatible weight vector is shown to be unique to the extent of a multiplicative constant, and its structure is specified explicitly. In order to apply this mechanism in very-large-scale P2P streaming network, we propose an adaptive algorithm for distributed computation of the incentive compatible discount vector. The simulation results show that the incentive compatible pricing mechanism can avoid the routing hotspot efficiently.