A Hitchhiker's Guide to Computation Offloading: Opinions from Practitioners
Morteza Golkarifard, Ji Yang, Ali Movaghar, Pan Hui · IEEE Communications Magazine · 2017
Due to the increasing usage and capabilities of smart devices, mobile application developers build a large number of resource intensive applications, such as WAR applications. Even with the rapid development in hardware technology, the computing capability and battery capacity on wearable devices and smartphones still cannot meet the application demands with heavy computations and high battery drain. Pervasive computing addresses this problem by migrating applications to the resource providers external to mobile devices. The profitability of this method heavily depends on how to implement it and when to use it. Although there are many computation offloading systems proposed in the literature, there is no practical manual that addresses all the implementation complexities on the way to building a general offloading system. In this article, we review developments in the field of pervasive computing on computation offloading. We use this literature review together with our own experience and provide designers with some detailed guidelines to gain a deep insight into the implementation challenges of a computation offloading system. The guidelines empower the reader to choose between the variety of solutions in the literature for developing any offloading system with the consideration of their own system architecture and available facilities. Finally, we evaluate our general offloading system on Android devices with two real-time applications.