Infrastructureless storage in dynamic environments

This paper studies the use of highly dynamic networks as infrastructures for persistent storage of data that offer services at specific geographical zones in a decentralized and distributed way. We propose a new algorithm, based on repulsion techniques, to self-organize the nodes that store and serve the information. In this work, we focus on the evaluation of our algorithm when faced to different simulated failures in order to measure its robustness and compare it with an existing approach.

[1]  Patrick Th. Eugster,et al.  Location-based Publish/Subscribe , 2005, Fourth IEEE International Symposium on Network Computing and Applications.

[2]  Giovanna Di Marzo Serugendo,et al.  Hovering Information - Self-Organising Information that Finds Its Own Storage , 2008, 2008 IEEE International Conference on Sensor Networks, Ubiquitous, and Trustworthy Computing (sutc 2008).

[3]  Cecilia Mascolo,et al.  Persistent content-based information dissemination in hybrid vehicular networks , 2009, 2009 IEEE International Conference on Pervasive Computing and Communications.

[4]  Karl Aberer,et al.  Autonomous Gossiping: A Self-Organizing Epidemic Algorithm for Selective Information Dissemination in Wireless Mobile Ad-Hoc Networks , 2004, ICSNW.

[5]  Sándor P. Fekete,et al.  Recognizing Traffic Jams with Hovering Data Clouds , 2006, Second International Symposium on Leveraging Applications of Formal Methods, Verification and Validation (isola 2006).

[6]  Jose Luis Fernandez-Marquez,et al.  Keeping diversity when exploring dynamic environments , 2009, SAC '09.

[7]  Cecilia Mascolo,et al.  Epcast: Controlled Dissemination in Human-based Wireless Networks by means of Epidemic Spreading Models , 2007, ArXiv.

[8]  I. Leontiadis,et al.  Opportunistic spatio-temporal dissemination system for vehicular networks , 2007, MobiOpp '07.

[9]  Radhika Nagpal,et al.  Robust and Self-Repairing Formation Control for Swarms of Mobile Agents , 2005, AAAI.