Point-to-Point and Congestion Bandwidth Estimation: Experimental Evaluation on PlanetLab Data

In large scale Internet platforms, measuring the available bandwidth between nodes of the platform is difficult and costly. However, resource optimization for large-scale Internet applications (like broadcasting a message or organizing master/slave computations) requires to have predictions about the performance of the platform, and especially about available bandwidth. In this paper, we analyze the feasibility to provide estimations, based on a limited number of pairwise measurements, for the point-to-point available bandwidth values, and for the congestion which happens when several communications take place at the same time. We present a dataset containing both types of measurements performed on a set of nodes from the PlanetLab platform. We study two estimation methods, namely matrix factorization and the LastMile model. We show that matrix factorization techniques are quite efficient at predicting point-to-point available bandwidth, but are not adapted for congestion analysis. However, a LastMile modeling of the platform allows to perform congestion predictions with a reasonable level of accuracy, even with a small amount of information, despite the variability of the measured platform.