Paper Info
Title
Computational offloading mechanism for native and android runtime based mobile applications.
Abstract
We empirically evaluate the current state of the computational offloading categories.We identified research gap in computational offloading with the introduction of ART.We propose an offloading framework for native and ART based mobile application.We provide the proof of concept experiment which validates the proposed framework. Mobile cloud computing is a promising approach to augment the computational capabilities of mobile devices for emerging resource-hungry mobile applications. Android-based smartphones have opened real-world venues for mobile cloud applications mainly because of the open source nature of Android. Computational offloading mechanism enables the augmentation of smartphone capabilities. The problem is majority of existing computational offloading solutions for Android-based smartphones heavily depends on Dalvik VM (an application-level VM). Apart from being a discontinued product, Dalvik VM consumes extra time and energy because of the just-in-time (JIT) compilation of bytecode into machine instructions. With regard to this problem, Google has introduced Android Runtime (ART) featuring ahead-of-time (AHOT) compilation to native instructions in place of Dalvik VM. However, current state-of-the-art offloading solutions do not consider AHOT compilations to native binaries in the ART environment. To address the issue in offloading ART-based mobile applications, we propose a computational offloading framework. The proposed framework requires infrastructural support from cloud data centers to provide offloading as a service for heterogeneous mobile devices. Numerical results from proof-of-concept implementation revealed that the proposed framework improves the execution time of the experimental application by 76% and reduces its energy consumption by 70%.
Year
DOI
Venue
2016
10.1016/j.jss.2016.07.043
Journal of Systems and Software
Keywords
Field
DocType
Mobile cloud computing,Computational offloading,Android runtime environment,Application partitioning,Mobile cloud scheduling
Mobile cloud computing,Android (operating system),Computer science,Real-time computing,Mobile device,Proof of concept,Execution time,Bytecode,Energy consumption,Operating system,Embedded system,Cloud computing
Journal
Volume
Issue
ISSN
121
C
0164-1212
Citations 
PageRank 
References 
6
0.49
21
Authors
6
Name
Order
Citations
PageRank
Abdullah Yousafzai160.49
Abdullah Gani2188791.22
Rafidah Md Noor325530.34
Anjum Naveed4979.13
Raja Wasim Ahmad517810.30
Victor Chang61202107.48