A comprehensive data management framework for opportunistic communication on mobile phones

ABSTRACT

Several of our non-profit partners working in rural areas complain about poor data connectivity from their mobile phones. To instrument this, we deployed a simple application on Android mobile phones of two field staff located in the state of Jharkhand in India to continuously probe 2G GPRS EDGE connectivity across several days. We found that the connectivity was quite flaky and underwent frequent disruptions as the staff moved around for their work. This motivated us to develop a comprehensive data management framework that can run on mobile devices and help application developers cope with several issues including communication on flaky connections, data synchronization, support for transactions, and consistency management. Most previous work in the area of supporting communication in poorly connected regions has focused on connection management and session persistence across disconnections, while we focus more on data management challenges that arise in these scenarios. We have built and deployed an application for media transfer using this framework, and are now using this experience to improve the framework.

Our connectivity-testing Android application logged signal strengths and HTTP ping latencies to www.google.com to check for connection availability, and uploaded the traces every few hours to our server for analysis. The application was deployed on Samsung Galaxy Fit phones provided by us to two staff working with our field partner. Figure 1 shows the HTTP ping latencies (~ 2RTTs) and availability plotted on the map of Ranchi, the main city in which the field staff are located. The points in red indicate no connectivity, green points indicate moderate latency and the blue ones indicate high latency values. As can be seen, the mobile devices often run into areas of poor availability, and we found the mean time between disconnections and the maximum disconnection period to be 83mins and 30mins respectively.

References

1. S. Keshav, "Design Principles for Robust Opportunistic Communication", NSDR, 2010
2. A. Seth, M. Zaharia, S. Keshav, and S. Bhattacharya, "A Policy-Oriented Architecture for Opportunistic Communication on Multiple Wireless Networks", Technical report, University of Waterloo, 2005
3. K. Fall, "A Delay Tolerant Network Architecture for Challenged Internets", SIGCOMM, 2003
4. D. B. Terry, M. M. Theimer, and K. Petersen, "Managing Update Conflicts in Bayou, a Weakly Connected Replicated Storage System", SOSP, 1995.
5. D. L. Johnson, V. Pejovic, E. M. Belding and G. V. Stam, "VillageShare: Facilitating content generation and sharing in rural networks", ACM DEV, 2012
6. S. Guo, M. Derakhshani, M. H. Falaki, U. Ismail, R. Luk, E. A. Oliver, S. Ur Rahman, A. Seth, M. A. Zaharia, and S. Keshav, "Design and Implementation of the KioskNet System", Computer Networks, 2011
7. A. Pentland, R. Fletcher and A. Hasson, "DakNet:Rethinking Connectivity in Developing Nations", IEEE Computer, 2004
8. A. Mahla, D. Martin, I. Ahuja, Q. Niyaz and A. Seth, "Motivation and Design of a Content Distribution Architecture for Rural Areas", ACM DEV, 2012
9. M. Demmer, B. Du and S. Surana, "TierStore: A Distributed Storage System for Developing Regions", FAST, 2008

Index Terms

• Applied computing

  • Computers in other domains

    • Personal computers and PC applications

• Information systems

  • Information systems applications