Context-Aware Mobile Application Task Offloading to the Cloud

Abstract

Abstract-One of the benefits of mobile cloud computing is the ability to offload mobile applications to the cloud for many reasons including performance enhancement and reduced resource consumption. This paper is concerned with offloading of context-aware mobile applications, in which actions or tasks are executed in certain contexts and offloading those tasks needs to be itself context-aware to be advantageous. The paper investigates candidate techniques and development models in the literature to identify suitable ones. Accordingly, the paper proposes the practical Context-Aware Mobile applications Offloading (CAMO) development model, which we developed in Java for the Android platform. Programmers can exploit the independency of the tasks of a typical context-aware mobile application and use CAMO to profile each task in isolation on the mobile and the cloud. The paper introduces the concept of a taskoffloading plan in which programmers specify a criterion and/or an objective for offloading a task in a specific context. Offloading criteria allow rapid offloading in case the mobile environment does not change frequently. Based on the profiling results, programmers can use the classes and methods of CAMO to develop one or more custom offloading plans for each task or use pre-specified plans, criterion and objectives. We provide three example tasks with details of their profiling and analysis for developing corresponding offloading plans. CAMO is general and flexible enough for offloading any application partitioned into independent modules. Empirical evaluation shows extreme satisfaction of mobile application developers with its capabilities.

Comparative Analysis of Network Libraries for Offloading Efficiency in Mobile Cloud Environment

“…Broadly, an offloading decision is made by utilizing the local information of the smartphones, namely, CPU and memory utilization, code profiling, network speed and/or user behavior. These parameters are fed into an optimization engine to take optimal decision to achieve the offloading objectives [6], [7], [8], [9]. The optimization decision can be either be taken on mobile device [10], cloud [9] or both [11].…”