Peer Me Maybe?: A Data-Centric Approach to ISP Peer Selection

The Internet landscape is progressively transitioning towards a flat hierarchical model to prune multiple Internet Service Provider (ISP) tiers. At the core of this transition is settlement free peering, which plays a critical role in mediating traffic exchange among ISPs. It is pertinent to take a closer look at peering and accurately emulate their operative model into a computation model that enables a concrete characterization without losing generality. In this paper, we utilize publicly available data-sets to identify the importance of several factors that play role in the peering process. We conduct a detailed analysis on the relationship of ISPs and their motivation behind selecting a peer ISP and use these findings to develop a Machine Learning (ML) based model that identifies feasible peering relationships. Preliminary results show a high correlation to the ground truth.