Visit ShanghaiTech University | 中文 | How to find us
HOME > News and Events > Events
User-Stash: A Novel System for Minimizing Mobile Data Download Costs and Augmenting the Quality of Experience
Date: 2014/11/17             Browse: 794

Speaker:Aruna Seneviratne

Time: Nov. 17, 2:00-3:30pm

Location: Room 220, Building 8, Yueyang Road Campus


As result of the increasing capabilities of mobile devices and the popularity of rich media content sharing services, mobile data traffic has been growing exponentially and users are struggling to keep within their download quotas and also experiences large startup delays. In this talk, describes a novel crowd-sourced local content storage system, referred to as user-stash, which provides a best effort, yet cheap and convenient content access for mobile users. User-stash achieves this by exploiting both the transient colocation of devices and the epidemic nature of content popularity, where users in a particular location at specific times are likely to be interested in the same set of content. In addition it provides an incentive mechanism based on in-app advertisements for the local store providers that can easily be incorporated with the existing advertisement eco system. The viability of user stash is demonstrated by modeling the system probabilistically, when used on public buses using data request and consumption characteristics derived from a large real-life video access dataset. We show that, in this particular setting, it is possible to serve as much as 60% the data locally within a one hour period and under any content popularity distribution and in general will be possible to serve a minimum of 40% content locally. Furthermore, that by using user stash leads to a reduction in the energy consumption of the mobile devices. The talk will conclude with a demonstration of the user stash system.


Aruna Seneviratne received his PhD in electrical engineering from the University of Bath , UK, in 1982. He is currently the leader of the Networks Research Group, and Director of the NICTA Australian Technology Park Laboratory, in Sydney, Australia. His research interests are in performance in mobile computing systems. He has held academic appointments at the University of Bradford, UK, Curtin University, University of Technology, Sydney and the University of New South Wales in Australia.  He has also held visiting appointments at the University of Pierre Marie Curie, Paris, and INRIA, Nice in France. In addition he has been consultant to numerous organizations, including Telstra, Vodafone, Inmarsat and Ericsson.

                                                                                                SIST-Seminar 14039