Why not just use class C subnets, AS numbers or measurement-based techiques for figuring out peer locality?
While heuristic-based approaches such as class C subnets and AS numbers also scale well, the position information gained through them is not terribly useful and does not take into account dynamic network conditions. For example, there may not be anyone downloading the same file on your class-C subnet and your AS number may correspond to a network that spans an entire country. Ono, on the other hand, finds peers that are near one another by relying on preexisting infrastucture (CDNs) that perform extensive Internet measurements. We have found that our system locates peers along paths that have two orders of magnitude lower latency and 30% lower loss rates than those picked at random by BitTorrent, and that these high-quality paths can lead to significant improvements in transfer rates.
