Network Datasets: what exists, and what are the problems?
Timur Friedman – Associate Professor, Sorbonne Université – LIP6
Phillipa Gill – Google, USA
Sue B. Moon – KAIST Chair Professor, South Korea
Dave Clark – MIT, USA
Italo Cunha – UFMG, Brazil
Christophe Diot – Principal Engineer at Network Operations team, Google
Jim Kurose – Professor of College of Information and Computer Sciences at the University of Massachusetts Amherst
Public datasets of network measurements have been created and made available for several decades. These datasets are interest to networking researchers (who are interested in workloads, topologies, and other characteristics of deployed networks in/across the backbone, access, home and mobile networks), students who want to learn what “real” networks look, policy makers, and more.
This NetworkingChannel event has two parts. The first part, we’ll identify and discuss public datasets of interest and their use. In the second part, we’ll identify some of the challenges of working with such datasets (including the difficulty of analyzing/comparing data longitudinally, the “ageing” of data) and the challenges of obtaining industry data (which may have significant commercial and proprietary value) and solutions to that challenge, such as benchmarks and workload models.
(11am EST / 5pm CET / 1am JST)
web-streamed | time streamed