PROJECT TITLE :

Rating Network Paths for Locality-Aware Overlay Construction and Routing

ABSTRACT:

This paper investigates the rating of network methods, i.e., acquiring quantized measures of path properties such as spherical-trip time and offered bandwidth. Compared to fine-grained measurements, coarse-grained ratings are appealing in that they're not solely informative but additionally low cost to obtain. Motivated by this insight, we initial address the scalable acquisition of path ratings by statistical inference. By observing similarities to recommender systems, we have a tendency to examine the applicability of solutions to a recommender system and show that our inference drawback will be solved by a class of matrix factorization techniques. A technical contribution is an active and progressive inference framework that not solely improves the accuracy by selectively measuring more informative methods, but additionally hurries up the convergence for obtainable bandwidth by incorporating its measurement methodology. Then, we tend to investigate the usability of rating-based network measurement and inference in applications. A case study is performed on whether or not locality awareness can be achieved for overlay networks of Pastry and BitTorrent using inferred ratings. We have a tendency to show that such coarse-grained information will improve the performance of peer choice which finer granularities don't forever lead to larger enhancements.