Abstract

This talk is focused on content distribution in community scenarios based on social interaction. Firstly, I will start analyzing the characteristics of data transfer based on swarms. Secondly I will introduce the factor of social similarities between peers. The result of this two perspectives will follow with the presentation of Swarmella, a framework for content distribution using delivery techniques based on swarming and social aspects. These functionalities are exploited by the Ranking Algorithm that manage the information a peer can collect from its neighbors found on each swarm and it is used to create the structure of the communities. We have evaluated SwarmTella through simulation and compared it with Gnutella. Our results show that SwarmTella clearly outperforms Gnutella, increasing the probability of success in the search procedure and reducing the bandwidth utilization in the search procedure. The current work is to extend SwarmTella in a delay-tolerant environments and validate it with real traces scenarios.

This is a joint work with Ruben Cuevas (UC3M)

This research is supported by: CONTENT EU NoE (www.ist-content.eu, BIOGRIDNET Project (www.biogridnet.es), IMDEA Networks Research Institute (www.imdea.org/networks) and NETCOM Research Group at University Carlos III Madrid (www.it.uc3m.es/netcom)

Short Bio: Carmen Guerrero received the Telecommunication Engineering degree in 1994 from the Technical University of Madrid (UPM), Spain, and the Ph.D. in Computer Science in 1998 from the Universidade da Coruña (UDC), Spain. She is currently associate professor since 2003 at Universidad Carlos III de Madrid (UC3M), teaching computer networks courses. She has been involved in several national and international research projects related with content distribution, overlay networks, information retrieval, broadband access networks, network management and advanced network and multimedia real time systems. More details of her profile are available at: www.it.uc3m.es/carmen