Feb 21st, 2014, 3:00 pm in Huxley 345

EdgeReduce: Eliminating Mobile Network Traffic Using Application-Specific Edge Proxies

This talk describes an approach for reducing network traffic over radio access networks (RANs), e.g. 3G and 4G networks, associated with mobile client applications such as social networking, photo sharing and e-commerce clients. Such applications rely on frequent interactions with Internet backend services, which often entail unnecessary data transfers due to the coarse granularity of backend API calls and the aggressive prefetching strategies used. Our approach automatically partitions mobile client applications to offload the application logic responsible for processing API response data to remote nodes at edge locations of a mobile network. This allows for unused data to be discarded before traversing the RAN. This benefits both mobile users—by reducing increased data usage charges—and network operators—by reducing network contention in limited RANs.

Andreas Pamboris is currently a post-doc at Imperial College London, where he is a member of the Large Scale Distributed Systems (LSDS) group. He completed his undergraduate studies at the Computer Science department of the University of Cyprus, followed by a Masters degree at the Computer Science and Engineering department of the University of California, San Diego. During this period of time he has done research on wireless sensor networks and data-center networking. Andreas recently completed his PhD at Imperial College London under the supervision of Dr. Peter Pietzuch. His research during the past 4 years has focused on mobile cloud computing solutions. More specifically, his work delves into automatic code offloading approaches that leverage cloud-based resources to enhance the capabilities of resource-constrained devices via static application partitioning techniques, which allow mobile applications to execute across mobile devices and the cloud with increased performance and functionality. More about me→