Promethee Spathis

Promethee Spathis is an Associate Professor in the Department of Computer Science at Sorbonne University, where he has been since 2005. From 2015 to 2018, he was a Visiting Associate Professor at New York University (Shanghai Campus) during the Spring terms. He received a B.S. in 1997, and an M.S. in 1999 from Université Pierre et Marie Curie (UPMC) in France. He received his Ph.D. in Computer Science in 2004 and his HDR (Habilitation à Diriger des Recherches) in 2016 from Université Pierre et Marie Curie.

His research interests span both computer networking and network science. Much of his work has been on improving the understanding, design, and performance of mobile networked systems, mainly through the application of data mining and performance evaluation. He has made numerous contributions to mobile computing in the context of challenging environments and disruptive usage patterns. He has examined the impact of mobility patterns on the design of data offloading systems as alternative or complement to conventional infrastructure-based data networks.

Dr Promethee Spathis was and is still leading or involved in many research projects in High Performance Networking in France, Europe, and South America. He has been involved in the FP7 MOTO and FP6 ANA IP projects, the ACCA coordination action, and many national research projects as a participant or the scientific coordinator including ANR DataTweet, ANR CONNECT, and ANR RESCUE. He has also participated to three French scientific-technological cooperation programmes with South America.

The number and storage capacity of the handled devices we carry with us everywhere has significantly increased. In this presentation, we present three pieces of work where we exploit the opportunity of transferring data stored on such devices as their owners move according to their daily commute. In a first work, we present a transmission system that takes advantage of the daily trips made by car to offload large volumes of data from the Internet. Our system extends cost effectively the capacity of the Internet by having cars transporting data over long distances. In a second work, we present a stand-alone cloud storage system targeting mobile users in urban environments. This system relies on a collection of strategically located wireless data storage equipments acting as file repositories where mobile users can upload and share their files to save storage space on their devices. In a last work, we address the problem of virtualizing the resources of large-scale mobile networks. This work focuses on the migration of virtual machines triggered by the reallocations of virtual resources or changes in the physical topology.