Scalable and systematic Internet-wide path and delay estimation from existing measurements

The main contribution of this paper is an estimation algorithm that breaks down measurement data into segments, efficiently identifies relevant segments, and, by carefully stitching segments together, produces delay and path estimates between any two end points. Fittingly, we call our algorithm path stitching. Our results show remarkably good accuracy: error in delay is below 20 ms in 80% of end-to-end paths. We also show that our path stitching approach performs comparably to existing iPlane without having to instrument any new measurement node.