Poster Abstract: ContextMete: Context Optimized Peer Offload Architecture for Distributed Mobile Applications

Madhurima Pore, Vinaya Chakati, Ayan Banerjee, Sandeep K. S. Gupta

Abstract: Increased availability of wearable devices, smart phones establishes a platform for designing innovative distributed crowd sourcing mobile applications. These applications can take advantage of computation offloading to augment the constrained resources and unavailability of required context on single mobile device. ContextMete is an offloading architecture that execute such applications across the optimized set of nearby volunteer devices that have the required context. It uses human mobile interaction models to predict context on the mobile for context optimization in offload decisions. ContextMete is evaluated using construction of 360° video using dynamically discovered devices situated in four strategic locations of the stadium.