To offload or not to offload? The bandwidth and energy costs of mobile cloud computing

Abstract

The cloud seems to be an excellent companion of mobile systems, to alleviate battery consumption on smartphones and to backup user's data on-the-fly. Indeed, many recent works focus on frameworks that enable mobile computation offloading to software clones of smartphones on the cloud and on designing cloud-based backup systems for the data stored in our devices. Both mobile computation offloading and data backup involve communication between the real devices and the cloud. This communication does certainly not come for free. It costs in terms of bandwidth (the traffic overhead to communicate with the cloud) and in terms of energy (computation and use of network interfaces on the device).In this work we study the feasibility of both mobile computation offloading and mobile software/data backups in real-life scenarios. In our study we assume an architecture where each real device is associated to a software clone on the cloud. We consider two types of clones: The off-clone, whose purpose is to support computation offloading, and the back-clone, which comes to use when a restore of user's data and apps is needed. We give a precise evaluation of the feasibility and costs of both off-clones and back-clones in terms of bandwidth and energy consumption on the real device. We achieve this through measurements done on a real testbed of 11 Android smartphones and an equal number of software clones running on the Amazon EC2 public cloud. The smartphones have been used as the primary mobile by the participants for the whole experiment duration. I. INTRODUCTIONThe advances in technology of the last decades have undoubtedly turned yesterday's must-have devices into today's stock. Think of the phones with aerials of the late '80, or the Pentium 4 PCs of a few years ago. None of them is comparable to the power of nowadays smartphones, whose recent worldwide market boost is undeniable. We use smartphones to do many of the jobs we used to do on desktops, and many new ones. We browse the Internet, send emails, organize our lives, watch videos, upload data on social networks, use online banking, find our way by using GPS and online maps, and communicate in revolutionary ways. New apps are coming out at an incredible pace. Apple iPhone commercial's call to action "There's an app for everything" says a lot on this Alessandro Mei is supported by a Marie Curie Outgoing International Fellowship funded by the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n. 253461. This work has been technically supported and partially funded by Telecom Italia within the Working Capital project.This work has been performed in the framework of the FP7 project TROPIC IST-318784 STP, which is funded by the European Community. The Authors would like to acknowledge the contributions of their colleagues from TROPIC Consortium (http://www.ict-tropic.eu).

The Cloud as an OpenMP Offloading Device

“…Contrary to our approach which relies on C/C++, they mostly rely on .NET or Java which are the most popular environments for mobile device. One of the key feature of those offloading framework, not treated in this paper, is to dynamically determine if it worth offloading the computation in term of communication overhead and energy consumption [33]. Some of those frameworks even explore the parallelization of the execution by providing multi-threading support, or virtual machine duplication, but it requires a considerable programming effort for non programming expert and their result do not present very large speedups (up to 4x).…”

Power-aware anomaly detection in smartphones: An analysis of on-platform versus externalized operation

“…ParanoidAndroid [29], Secloud [42] and CloudShield [1] are illustrative examples of such systems. In these cases, all security-related tasks, including monitoring, analysis and detection can be performed in an environment not exposed to battery constraints.…”

Green Cloud Computing for Multi Cell Networks

“…Interference and user's QoS is neglected in this study. In [11], it is shown that wireless access has an inevitable effect on the performance of MCC, using experimental measurements. The authors in [12], consider the problem of resource scheduling for multi service multi user MCC networks.…”