Partitioning Android mobile application for improving execution time

Smartphones with improved technologies and features have enabled development of sophisticated mobile applications. Still, many mobile applications do not perform well due to the battery capacity and shortage of resources for computation. These limitations can be overcome through computing offloading i.e. sending heavy computation to resourceful devices and getting results. We propose a way of executing android applications on a local cloud formed by devices in vicinity. The local cloud is created using Wi-Fi Direct technology. We have also proposed a method for dividing computation task and migrating it to mobile local cloud. The project implements a mobile local cloud for increasing, augmenting, and optimizing computation capabilities of mobile devices which aims to execute applications that are resource-intensive. The workload is migrated on system and application levels according to available device resources such as network connectivity and computation capacity. The ubiquitous nature and flexibility of android makes this idea feasible. We implement a test application to support our concept and show evaluation outcomes.