Framework for Adaptive Computation Offloading in IoT
Applications
Shihong Chen
1,2
, Bichun Liu
1,2
, Xing Chen
1,2+
, Ying Zhang
3,4
, Gang Huang
3,4
1
College of Mathematics and Computer Science, Fuzhou University, Fuzhou, China
2
Fujian Key Laboratory of Network Computing and Intelligent Information Processing, Fuzhou, China
3
Institute of Software, School of Electronics Engineering and Computer Science, Peking University, Beijing, China
4
Key Laboratory of High Confidence Software Technologies, Ministry of Education, Beijing, China
+ Corresponding author: chenxing@fzu.edu.cn
Abstract: The internet of things (IoT) attracts great interest in
many application domains concerned with monitoring and
control of physical phenomena. IoT applications try to provide
more and more functionality and then they inevitably become
so complex as to make the limits of devices worse, which may
lead to poor performance of applications. Computation
offloading is a promising way to improve the performance of an
IoT application by executing some parts of the application on
remote devices or servers. However, supporting such capability
is not easy for application developers due to (1) adaptability: IoT
applications often face changes of runtime environments so
that the adaptation on offloading is needed. (2) effectiveness:
when the device context changes, it needs to dynamically
decide the deployment plan of computation tasks, and the
reduced execution time must be greater than the network delay
and extra overheads caused by offloading. This paper proposes
a framework which supports IoT applications with adaptive
computation offloading capability. First, a design pattern is
proposed to enable an application to be computation offloaded
on-demand. Second, an estimation model is presented to
automatically decide the deployment plan for offloading. Third,
a framework is implemented to support the design pattern and
the estimation model. A thorough evaluation on the real-world
application is proposed, and the results show that our approach
can help reduce execution time by over 45% in most scenarios.
Keywords:Internet of Things; Computation Offloading;
Adaptive Software.
Permission to make digital or hard copies of all or part of this work for personal
or classroom use is granted without fee provided that copies are not made or
distributed for prot or commercial advantage and that copies bear this notice and
the full citation on the rst page. Copyrights for components of this work owned
by others than ACM must be honored. Abstracting with credit is permied. To
copy otherwise, or republish, to post on servers or to redistribute to lists, requires
prior specic permission and/or a fee. Request permissions from
Permissions@acm.org.
Internetware'17, September 23, 2017, Shanghai, China
© 2017 Association for Computing Machinery.
ACM ISBN 978-1-4503-5313-7/17/09…$15.00
hps://doi.org/10.1145/3131704.3131717
1 Introduction
The internet of things (IoT) is a novel paradigm that the
pervasive presence of a variety of inter-networked objects
(things) are able to interact and cooperate with each other to
reach common goals [1]. Thousands of devices and sensors are
spread in specific areas to collect all kinds of physical
information and connected to the Internet. They are attracting
great interest in many application domains concerned with
monitoring and control of physical phenomena. Following the
fast improvement of IoT hardware and increased user
experience, applications try to provide more and more
functionality and then they inevitably become so complex as to
make the two most critical limits of IoT infrastructure worse.
The first limit is the network connections. Complex
applications usually require a great deal of data traffic [2], such
as audio, video and image data. For instance, a HD-capable
camera can capture 42 GB video data per day. Although the
network connection keeps growing continuously, it still cannot
keep pace with the growing requirements of IoT applications
[3].
The second limit is the diversity of hardware configurations.
Some computation tasks have been deployed and executed on
IoT devices, in order to reduce the data traffic. Thus, the
diversity of hardware configurations may give users very
different experiences even running the same applications [4].
Generally speaking, the low hardware configuration of a IoT
device implies the low performance of the application running
on the device, and gives a poor experience to the user.
Computation offloading can be a promising technique to
help improve the performance of an IoT application [5][6][7].
Offloading, also referred to as remote execution, is to make
some computation intensive code of an application executed on
a remote server, so that the application can take advantage of
the powerful hardware for increasing its responsiveness.
Computation offloading has been successfully applied to
mobile cloud computing, assuming that mobile codes are
offloaded to one server or one Cloud [8][9][10][11]. However,
the IoT application is usually deployed in a distributed,
heterogeneous and dynamic environment. With respect of
features of IoT infrastructures, two advanced issues have to be
dealt with:
Adaptability: the IoT application is deployed on multiple
devices and servers, and it often faces changes of runtime
environments, so that the adaptation on offloading is needed. It
资源评论
